feature-requirements-clarification
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (feature ideas and chat records) to generate and update project documentation.
- Ingestion points: User input is accepted during the initial phase defined in the Input section of
SKILL.md. - Boundary markers: The skill refers to
specs/GUARDRAILS.md, but no specific delimiters are used to wrap user-provided strings when they are written to files. - Capability inventory: The agent can perform file writes to
docs/andspecs/directories (e.g.,1_需求文档.mdand1_产品概述.md). - Sanitization: There is no explicit sanitization or escaping of the user-provided input before it is interpolated into the generated documents.
Audit Metadata