hunt-api-misconfig
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted external data, including API responses, OData metadata, and OpenAPI specifications. This behavior creates a potential surface for indirect prompt injection if a malicious target embeds instructions in its responses. * Ingestion points: SKILL.md (Step 3 in Hunt sequence, OData detection heuristics, Swagger discovery paths). * Boundary markers: None identified. * Capability inventory: Shell command execution (curl), object manipulation, and data parsing. * Sanitization: No explicit sanitization of external API content is described.
- [COMMAND_EXECUTION]: The skill documents the use of standard security tools and provides payloads for demonstrating proof-of-concept impact. * Evidence: Use of curl for CORS testing and Node.js execSync for verifying prototype pollution sinks. These are provided as instructional examples.
Audit Metadata