Affected: Homebrew: aider

### Impact

When JWT authentication is enabled (`enable_jwt_auth: true`), the OIDC userinfo cache uses `token[:20]` as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters.

This configuration option is not enabled by default. **Most instances are not affected.**

An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled.

### Patches

Fixed in v1.83.0. The cache key now uses the full hash of the JWT token.

### Workarounds

Disable OIDC userinfo caching by setting the cache TTL to 0, or disable JWT authentication entirely.

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.83.0, when JWT authentication is enabled (enable_jwt_auth: true), the OIDC userinfo cache uses token[:20] as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters. This configuration option is not enabled by default. Most instances are not affected. An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled. Fixed in v1.83.0.

Affected: PyPI: litellm

### Impact

When JWT authentication is enabled (`enable_jwt_auth: true`), the OIDC userinfo cache uses `token[:20]` as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters.

This configuration option is not enabled by default. **Most instances are not affected.**

An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled.

### Patches

Fixed in v1.83.0. The cache key now uses the full hash of the JWT token.

### Workarounds

Disable OIDC userinfo caching by setting the cache TTL to 0, or disable JWT authentication entirely.

Affected: PyPI: litellm

### Impact

When JWT authentication is enabled (`enable_jwt_auth: true`), the OIDC userinfo cache uses `token[:20]` as the cache key. JWT headers produced by the same signing algorithm generate identical first 20 characters.

This configuration option is not enabled by default. **Most instances are not affected.**

An unauthenticated attacker can craft a token whose first 20 characters match a legitimate user's cached token. On cache hit, the attacker inherits the legitimate user's identity and permissions. This affects deployments with JWT/OIDC authentication enabled.

### Patches Fixed in v1.83.0. The cache key now uses the full hash of the JWT token.

### Workarounds

Disable OIDC userinfo caching by setting the cache TTL to 0, or disable JWT authentication entirely.

crypto/tls: crypto/tls: Incorrect certificate validation during TLS session resumption aiohttp: aiohttp: Denial of Service via specially crafted POST request aiohttp: aiohttp: Denial of Service via memory exhaustion from crafted POST request ansible-lightspeed: Broken Object Level Authorization Leading to Cross-User AI Conversation Context Injection in Ansible Lightspeed API lodash: lodash: Arbitrary code execution via untrusted input in template imports path-to-regexp: path-to-regexp: Denial of Service via crafted regular expressions aap-controller: aap-gateway: Account hijacking and unauthorized access via unverified email linking aap-gateway: missing requestHeadersToRemove allows mTLS bypass via Subject header spoofing pyasn1: pyasn1: Denial of Service due to memory exhaustion from malformed RELATIVE-OID wheel: wheel: Privilege Escalation or Arbitrary Code Execution via malicious wheel file unpacking net/url: Incorrect parsing of IPv6 host literals in net/url cryptography: cryptography Subgroup Attack Due to Missing Subgroup Validation for SECT Curves crypto/x509: Incorrect enforcement of email constraints in crypto/x509 pyOpenSSL: DTLS cookie callback buffer overflow rollup: Rollup: Remote Code Execution via Path Traversal Vulnerability svgo: SVGO: Denial of Service via XML entity expansion pyasn1: pyasn1 Vulnerable to Denial of Service via Unbounded Recursion quinn-proto: quinn-proto: Denial of Service via crafted QUIC Initial packet black: Black: Arbitrary file writes…