GHSA-PQ5P-34CR-23V9

Vulnerability from github – Published: 2025-10-10 20:26 – Updated: 2025-11-03 18:31
VLAI?
Summary
Authlib is vulnerable to Denial of Service via Oversized JOSE Segments
Details

Summary Authlib’s JOSE implementation accepts unbounded JWS/JWT header and signature segments. A remote attacker can craft a token whose base64url‑encoded header or signature spans hundreds of megabytes. During verification, Authlib decodes and parses the full input before it is rejected, driving CPU and memory consumption to hostile levels and enabling denial of service.

Impact

  • Attack vector: unauthenticated network attacker submits a malicious JWS/JWT.

  • Effect: base64 decode + JSON/crypto processing of huge buffers pegs CPU and allocates large amounts of RAM; a single request can exhaust service capacity.

  • Observed behaviour: on a test host, the legacy code verified a 500 MB header, consuming ~4 GB RSS and ~9 s CPU before failing.

  • Severity: High. CVSS v3.1: AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H (7.5).

Affected Versions Authlib ≤ 1.6.3 (and earlier) when verifying JWS/JWT tokens. Later snapshots with 256 KB header/signature limits are not affected.

Proof of concept

Local demo (do not run against third-party systems): Download jws_segment_dos_demo.py the PoC in direcotry authlib/ Run following Command

python3 jws_segment_dos_demo.py --variant both --sizes "500MB" --fork-per-case

Environment: Python 3.13.6, Authlib 1.6.4, Linux x86_64, CPUs=8 Sample output: Refined image

The compilation script prints separate “[ATTACKER]” (token construction) and “[SERVER]” (Authlib verification) RSS deltas so defenders can distinguish client-side preparation from server-side amplification. Regression tests authlib/tests/dos/test_jose_dos.py further capture the issue; the saved original_util.py/original_jws.py reproductions still accept the malicious payload.

Remediation

  • Apply the upstream patch that introduces decoded size limits:

  • MAX_HEADER_SEGMENT_BYTES = 256 KB

  • MAX_SIGNATURE_SEGMENT_BYTES = 256 KB

  • Enforce Limits in authlib/jose/util.extract_segment and _extract_signature.

  • Deploy the patched release immediately.

  • For additional defence in depth, reject JWS/JWT inputs above a few kilobytes at the proxy or WAF layer, and rate-limit verification endpoints.

Workarounds (temporary)

  • Enforce input size limits before handing tokens to Authlib.

  • Use application-level throttling to reduce amplification risk.

Resources

  • Demo script: jws_segment_dos_demo.py

  • Tests: authlib/tests/dos/test_jose_dos.py

  • OWASP JWT Cheat Sheet (DoS guidance)

Severity ?
7.5 (High)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Show details on source website
To clipboard 

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "authlib"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.6.5"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2025-61920"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-20",
      "CWE-400",
      "CWE-770"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2025-10-10T20:26:43Z",
    "nvd_published_at": "2025-10-10T20:15:37Z",
    "severity": "HIGH"
  },
  "details": "**Summary**\nAuthlib\u2019s JOSE implementation accepts unbounded JWS/JWT header and signature segments. A remote attacker can craft a token whose base64url\u2011encoded header or signature spans hundreds of megabytes. During verification, Authlib decodes and parses the full input before it is rejected, driving CPU and memory consumption to hostile levels and enabling denial of service.\n\n**Impact**\n\n- Attack vector: unauthenticated network attacker submits a malicious JWS/JWT.\n\n- Effect: base64 decode + JSON/crypto processing of huge buffers pegs CPU and allocates large amounts of RAM; a single request can exhaust service capacity.\n\n- Observed behaviour: on a test host, the legacy code verified a 500\u202fMB header, consuming ~4\u202fGB RSS and ~9\u202fs CPU before failing.\n\n- Severity: High. CVSS v3.1: AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H (7.5).\n\nAffected Versions\nAuthlib \u2264\u202f1.6.3 (and earlier) when verifying JWS/JWT tokens. Later snapshots with 256\u202fKB header/signature limits are not affected.\n\n**Proof of concept**\n\nLocal demo (do not run against third-party systems):\nDownload [jws_segment_dos_demo.py](https://github.com/user-attachments/files/22450820/jws_segment_dos_demo.py) the PoC in direcotry authlib/\nRun following Command\n```\npython3 jws_segment_dos_demo.py --variant both --sizes \"500MB\" --fork-per-case\n\n```\nEnvironment: Python 3.13.6, Authlib 1.6.4, Linux x86_64, CPUs=8 \nSample output: Refined\n\u003cimg width=\"1295\" height=\"306\" alt=\"image\" src=\"https://github.com/user-attachments/assets/6dd8410f-bc36-4717-8cee-649bac9bf291\" /\u003e\n\n\n\n\nThe compilation script prints separate \u201c[ATTACKER]\u201d (token construction) and \u201c[SERVER]\u201d (Authlib verification) RSS deltas so defenders can distinguish client-side preparation from server-side amplification. Regression tests authlib/tests/dos/test_jose_dos.py further capture the issue; the saved original_util.py/original_jws.py reproductions still accept the malicious payload.\n\n**Remediation**\n\n- Apply the upstream patch that introduces decoded size limits:\n\n- MAX_HEADER_SEGMENT_BYTES = 256 KB\n\n- MAX_SIGNATURE_SEGMENT_BYTES = 256 KB\n\n- Enforce Limits in authlib/jose/util.extract_segment and _extract_signature.\n\n- Deploy the patched release immediately.\n\n- For additional defence in depth, reject JWS/JWT inputs above a few kilobytes at the proxy or WAF layer, and rate-limit verification endpoints.\n\n**Workarounds (temporary)**\n\n- Enforce input size limits before handing tokens to Authlib.\n\n- Use application-level throttling to reduce amplification risk.\n\n**Resources**\n\n- Demo script: jws_segment_dos_demo.py\n\n- Tests: authlib/tests/dos/test_jose_dos.py\n\n- OWASP JWT Cheat Sheet (DoS guidance)",
  "id": "GHSA-pq5p-34cr-23v9",
  "modified": "2025-11-03T18:31:46Z",
  "published": "2025-10-10T20:26:43Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/authlib/authlib/security/advisories/GHSA-pq5p-34cr-23v9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-61920"
    },
    {
      "type": "WEB",
      "url": "https://github.com/authlib/authlib/commit/867e3f87b072347a1ae9cf6983cc8bbf88447e5e"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/authlib/authlib"
    },
    {
      "type": "WEB",
      "url": "https://lists.debian.org/debian-lts-announce/2025/10/msg00032.html"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Authlib is vulnerable to Denial of Service via Oversized JOSE Segments"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Sightings

Author Source Type Date

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.



Detection rules are retrieved from Rulezet.