auth-bypass

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs embedding JWTs, API tokens, cookies, and signing secrets verbatim into commands and headers (e.g., Authorization: Bearer TOKEN, -p SECRET, curl/cli examples), which requires the LLM to handle and output secret values directly and thus poses a high exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document provides explicit, actionable techniques to bypass authentication and authorization (JWT alg=none, key-confusion, JWKS spoofing, token brute-force, kid path traversal, cookie/IDOR manipulation and forced browsing), which enable unauthorized access, credential compromise, and data exfiltration — high risk for misuse.