requirement-clarify
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted requirement descriptions and chat logs which could contain malicious instructions.\n
- Ingestion points: The skill explicitly reads '需求来源方的原话或聊天记录' (original words or chat logs from the requirement source) as primary input.\n
- Boundary markers: The skill instructs the agent to use specific tags like [INFERRED], [UNKNOWN], and [自答] (self-answered) to distinguish between data types, and to label facts with their sources.\n
- Capability inventory: The agent is authorized to read the local codebase and documentation (docs/product/, docs/prd/) and write summary files to docs/prd//clarify.md.\n
- Sanitization: No technical sanitization of inputs is performed, but the structured interview format and requirement for source attribution mitigate accidental instruction following.\n- [NO_CODE]: The skill consists entirely of markdown instructions and metadata, with no accompanying scripts or executables.
Audit Metadata