hunt-graphql

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill includes multiple curl and GraphQL examples that embed Authorization headers, tokens, and a plaintext password in request bodies/headers, which encourages inserting secret values verbatim into commands or outputs and therefore creates an exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The document contains explicit, reproducible exploitation instructions (curl sequences, payloads, attacker tokens), techniques for unauthorized privilege persistence and data access (race/desync, IDOR, SSRF, PII enumeration), and tooling/automation tips that can be directly used to perform illicit access or exfiltration.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes a “hunt-graphql” methodology that uses HTTP POST requests to outsider-controlled targets’ /graphql endpoints with user-supplied query strings (e.g., introspection queries), meaning the runtime ingests attacker-authored free text as the GraphQL “query” body after endpoint discovery.