Action not permitted
Modal body text goes here.
Modal Title
Modal Body
GHSA-9PG5-8C7X-5HFX
Vulnerability from github – Published: 2026-10-30 14:54 – Updated: 2026-10-30 14:54Summary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "joserfc"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "1.7.6"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-119284"
],
"database_specific": {
"cwe_ids": [
"CWE-400"
],
"github_reviewed": true,
"github_reviewed_at": "2026-10-30T14:54:25Z",
"nvd_published_at": "2026-10-29T22:17:01Z",
"severity": "MODERATE"
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "GHSA-9pg5-8c7x-5hfx",
"modified": "2026-10-30T14:54:25Z",
"published": "2026-10-30T14:54:25Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.4.0",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection"
}
BREW-FASTMCP-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 10:19 – Updated: 2026-11-11 10:24 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.8.0",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "fastmcp",
"purl": "pkg:brew/fastmcp"
},
"ranges": [
{
"events": [
{
"introduced": "3.2.4_2"
},
{
"fixed": "4.1.2"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.8.0",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.8.0"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-fastmcp-CVE-2026-119284",
"modified": "2026-11-11T10:24:35Z",
"published": "2026-10-31T10:19:08Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-KASKADE-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 10:32 – Updated: 2026-10-31 10:32 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": null,
"range_state": "affected",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.5",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "kaskade",
"purl": "pkg:brew/kaskade"
},
"ranges": [
{
"events": [
{
"introduced": "4.0.7_4"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.5",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.5"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-kaskade-CVE-2026-119284",
"modified": "2026-10-31T10:32:38Z",
"published": "2026-10-31T10:32:38Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-KIMI-CLI-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 11:03 – Updated: 2026-10-31 11:03 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": null,
"range_state": "affected",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.5",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "kimi-cli",
"purl": "pkg:brew/kimi-cli"
},
"ranges": [
{
"events": [
{
"introduced": "1.37.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.5",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.5"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-kimi-cli-CVE-2026-119284",
"modified": "2026-10-31T11:03:21Z",
"published": "2026-10-31T11:03:21Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-MCP-ATLASSIAN-CVE-2… (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 09:19 – Updated: 2026-11-10 09:52 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.8.0",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "mcp-atlassian",
"purl": "pkg:brew/mcp-atlassian"
},
"ranges": [
{
"events": [
{
"introduced": "0.21.1_3"
},
{
"fixed": "0.23.1_2"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.8.0",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.8.0"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-mcp-atlassian-CVE-2026-119284",
"modified": "2026-11-10T09:52:21Z",
"published": "2026-10-31T09:19:49Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-MCPM-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 10:32 – Updated: 2026-10-31 10:32 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.7",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "mcpm",
"purl": "pkg:brew/mcpm"
},
"ranges": [
{
"events": [
{
"introduced": "2.14.0_4"
},
{
"fixed": "2.15.0_5"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.7",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.7"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-mcpm-CVE-2026-119284",
"modified": "2026-10-31T10:32:15Z",
"published": "2026-10-31T10:32:15Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-MOTO-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 11:23 – Updated: 2026-10-31 11:23 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.7",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "moto",
"purl": "pkg:brew/moto"
},
"ranges": [
{
"events": [
{
"introduced": "5.0.2"
},
{
"fixed": "5.2.5"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.7",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.7"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-moto-CVE-2026-119284",
"modified": "2026-10-31T11:23:21Z",
"published": "2026-10-31T11:23:21Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-OTERM-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 09:09 – Updated: 2026-10-31 09:09 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.7",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "oterm",
"purl": "pkg:brew/oterm"
},
"ranges": [
{
"events": [
{
"introduced": "0.15.0"
},
{
"fixed": "0.26.1"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.7",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.7"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-oterm-CVE-2026-119284",
"modified": "2026-10-31T09:09:54Z",
"published": "2026-10-31T09:09:54Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-SAFETY-CVE-2026-119284 (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 08:46 – Updated: 2026-10-31 08:46 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.7",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "safety",
"purl": "pkg:brew/safety"
},
"ranges": [
{
"events": [
{
"introduced": "3.7.0_11"
},
{
"fixed": "3.8.1_3"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.7",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.7"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-safety-CVE-2026-119284",
"modified": "2026-10-31T08:46:31Z",
"published": "2026-10-31T08:46:31Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
BREW-WEAVIATE-CLI-CVE-20… (GHSA-9PG5-8C7X-5HFX)
Vulnerability from osv_homebrew – Published: 2026-10-31 09:57 – Updated: 2026-10-31 09:57 – Source websiteSummary
joserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but
calls value.split(b".") before validating them. Python bytes.split() copies
the attacker-controlled segments, so the current guard allocates roughly one
complete payload before rejecting it.
Reproduction
.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py
The input is resident before a fork-isolated verifier runs, and every signature is invalid.
| Input | Peak RSS delta | Amplification | Result |
|---|---|---|---|
| 1 MiB | 1.11 MiB | 1.11× | size rejected |
| 4 MiB | 4.04 MiB | 1.01× | size rejected |
| 8 MiB | 8.04 MiB | 1.01× | size rejected |
| 16 MiB | 16.04 MiB | 1.00× | size rejected |
| 32 MiB | 31.74 MiB | 0.99× | size rejected |
| 64 MiB | 63.80 MiB | 1.00× | size rejected |
The unconstrained profile is the primary evidence: joserfc allocates an
additional attacker-sized object before producing its intended
ExceededSizeError. A separate baseline-plus-headroom stress test raises
MemoryError when that copy cannot fit, but this report does not characterize
an absolute RLIMIT_AS value as a realistic worker memory budget.
Security Argument
The issue is not that joserfc can prevent the transport layer from receiving a
large header. The issue is narrower: once bytes reach deserialize_compact(),
the library's existing size guard cannot fulfill its allocation-limiting purpose
because attacker-controlled segment copies are created first.
This differs from GHSA-frfh-8v73-gjg4, which concerned embedding already-loaded oversized segments in exception and log messages. Here, the current implementation itself creates an additional payload-sized allocation before its fixed size exception.
Recommended Fix
Check len(value) against a maximum compact-serialization size before
split(). If per-segment diagnostics are needed, locate delimiters with bounded
find() operations or memory views, validate offsets and lengths, and only then
materialize acceptable segments.
{
"affected": [
{
"ecosystem_specific": {
"fix": "bump",
"range_state": "fixed",
"resource": "joserfc",
"resource_purl": "pkg:pypi/joserfc@1.7.7",
"upstream_fixed_in": "1.7.6"
},
"package": {
"ecosystem": "Homebrew",
"name": "weaviate-cli",
"purl": "pkg:brew/weaviate-cli"
},
"ranges": [
{
"events": [
{
"introduced": "3.4.1"
},
{
"fixed": "3.4.2_1"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"database_specific": {
"confidence": "high",
"source": "matched",
"strategy": "registry",
"upstream_evidence": [
{
"ecosystem": "PyPI",
"key": "pkg:pypi/joserfc@1.7.7",
"name": "joserfc",
"resource": "joserfc",
"strategy": "registry",
"subject_version": "1.7.7"
}
]
},
"details": "## Summary\n\njoserfc 1.7.4 defines compact-JWS header, payload, and signature limits, but\ncalls `value.split(b\".\")` before validating them. Python `bytes.split()` copies\nthe attacker-controlled segments, so the current guard allocates roughly one\ncomplete payload before rejecting it.\n\n## Reproduction\n\n```bash\n.venv-research/bin/python 06-practical-testing/preauth-memory-profile.py\n```\n\nThe input is resident before a fork-isolated verifier runs, and every signature\nis invalid.\n\n| Input | Peak RSS delta | Amplification | Result |\n|---:|---:|---:|---|\n| 1 MiB | 1.11 MiB | 1.11\u00d7 | size rejected |\n| 4 MiB | 4.04 MiB | 1.01\u00d7 | size rejected |\n| 8 MiB | 8.04 MiB | 1.01\u00d7 | size rejected |\n| 16 MiB | 16.04 MiB | 1.00\u00d7 | size rejected |\n| 32 MiB | 31.74 MiB | 0.99\u00d7 | size rejected |\n| 64 MiB | 63.80 MiB | 1.00\u00d7 | size rejected |\n\nThe unconstrained profile is the primary evidence: joserfc allocates an\nadditional attacker-sized object before producing its intended\n`ExceededSizeError`. A separate baseline-plus-headroom stress test raises\n`MemoryError` when that copy cannot fit, but this report does not characterize\nan absolute `RLIMIT_AS` value as a realistic worker memory budget.\n\n## Security Argument\n\nThe issue is not that joserfc can prevent the transport layer from receiving a\nlarge header. The issue is narrower: once bytes reach `deserialize_compact()`,\nthe library\u0027s existing size guard cannot fulfill its allocation-limiting purpose\nbecause attacker-controlled segment copies are created first.\n\nThis differs from\n[GHSA-frfh-8v73-gjg4](https://github.com/authlib/joserfc/security/advisories/GHSA-frfh-8v73-gjg4),\nwhich concerned embedding already-loaded oversized segments in exception and log\nmessages. Here, the current implementation itself creates an additional\npayload-sized allocation before its fixed size exception.\n\n## Recommended Fix\n\nCheck `len(value)` against a maximum compact-serialization size before\n`split()`. If per-segment diagnostics are needed, locate delimiters with bounded\n`find()` operations or memory views, validate offsets and lengths, and only then\nmaterialize acceptable segments.",
"id": "BREW-weaviate-cli-CVE-2026-119284",
"modified": "2026-10-31T09:57:52Z",
"published": "2026-10-31T09:57:52Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119284"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"type": "PACKAGE",
"url": "https://github.com/authlib/joserfc"
},
{
"type": "WEB",
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"schema_version": "1.7.3",
"severity": [
{
"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",
"type": "CVSS_V4"
}
],
"summary": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection",
"upstream": [
"GHSA-9pg5-8c7x-5hfx",
"CVE-2026-119284"
]
}
CVE-2026-119284 (GCVE-0-2026-119284)
Vulnerability from cvelistv5 – Published: 2026-10-29 21:33 – Updated: 2026-10-30 17:05- CWE-400 - Uncontrolled Resource Consumption
| URL | Tags |
|---|---|
| https://github.com/authlib/joserfc/security/advis… | x_refsource_CONFIRM |
| https://github.com/authlib/joserfc/commit/397e04f… | x_refsource_MISC |
| https://github.com/authlib/joserfc/releases/tag/1.7.6 | x_refsource_MISC |
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-119284",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-10-30T17:05:14.978930Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-10-30T17:05:38.184Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"references": [
{
"tags": [
"exploit"
],
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
}
],
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "joserfc",
"vendor": "authlib",
"versions": [
{
"status": "affected",
"version": "\u003c 1.7.6"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "joserfc is a Python library that provides an implementation of several JSON Object Signing and Encryption (JOSE) standards. Prior to 1.7.6, joserfc compact JWS deserialization calls bytes.split() before enforcing the existing aggregate segment-size limits. When attacker-controlled compact serialization bytes reach deserialize_compact(), the split creates copies of the header, payload, and signature segments before ExceededSizeError rejects them, causing an additional allocation approximately proportional to the supplied token. Requests that exhaust available process memory can therefore cause an availability failure even though signature verification fails. This issue is fixed in version 1.7.6."
}
],
"metrics": [
{
"cvssV4_0": {
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "NETWORK",
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"privilegesRequired": "NONE",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"vectorString": "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",
"version": "4.0",
"vulnAvailabilityImpact": "LOW",
"vulnConfidentialityImpact": "NONE",
"vulnIntegrityImpact": "NONE"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-400",
"description": "CWE-400: Uncontrolled Resource Consumption",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-10-29T21:33:16.529Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/authlib/joserfc/security/advisories/GHSA-9pg5-8c7x-5hfx"
},
{
"name": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/authlib/joserfc/commit/397e04f4bbce0b94d86e31c1c003f4640af8fb39"
},
{
"name": "https://github.com/authlib/joserfc/releases/tag/1.7.6",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/authlib/joserfc/releases/tag/1.7.6"
}
],
"source": {
"advisory": "GHSA-9pg5-8c7x-5hfx",
"discovery": "UNKNOWN"
},
"title": "joserfc: Compact JWS size guard allocates attacker-controlled segments before rejection"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-119284",
"datePublished": "2026-10-29T21:33:16.529Z",
"dateReserved": "2026-10-29T18:50:28.065Z",
"dateUpdated": "2026-10-30T17:05:38.184Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.