plan-interview
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external data from the codebase (such as README.md and AGENTS.md) and user-provided task descriptions to generate plans.
- Ingestion points: User inputs (task descriptions, interview responses) and local files (README.md, AGENTS.md).
- Boundary markers: Absent. The skill does not specify delimiters or "ignore" instructions for processed content.
- Capability inventory: Writing files to docs/plans/ and modifying task lists via TodoWrite.
- Sanitization: Absent. No validation or escaping of external content is performed.
- Mitigation: The risk is significantly reduced by the manual user review step required before any implementation begins.
- [SAFE]: No other security concerns such as hardcoded credentials, malicious network operations, or obfuscated code were detected in the instructions or reference materials.
Audit Metadata