api-negative-testing
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides static analysis and design guidance for API testing without requesting tool access, network capabilities, or file system modifications. It explicitly forbids the agent from executing tests or approving releases, ensuring human oversight at all steps.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted API documentation and raw responses. Ingestion points: User-provided API material analyzed according to prompts/api-negative-testing.md. Boundary markers: The instructions mandate a six-part input audit (known, missing, conflicting, stale, out_of_scope, assumptions) to categorize all ingested data. Capability inventory: None; the skill is strictly informational and prohibits all side-effect-causing actions in SKILL.md. Sanitization: The agent is required to separate facts, inferences, and recommendations, which logically isolates untrusted data from the agent's reasoning process.
Audit Metadata