ask-then-build
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill handles user-provided feature descriptions and answers to update documentation and generate prompts for other agents.
- Ingestion points: User input for feature ideas and clarifying questions defined in Phase 1.
- Boundary markers: No specific delimiters or safety warnings are specified to separate user-provided content from the agent's instructions.
- Capability inventory: The skill includes instructions to update local repository files (e.g., README, ADR, product requirements) with user-provided information.
- Sanitization: No explicit sanitization or validation logic is defined for the user input before it is written to files or used in prompt generation.
- [SAFE]: The skill operates entirely on local context and standard documentation practices. It does not perform network operations, execute remote code, or access sensitive credentials. It includes a protective rule ('don't commit') to prevent automated changes to the version history without manual review.
Audit Metadata