architecture-planning
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted user architectural requirements and persists them to a local file.
- Ingestion points: Architectural requirements provided by the user and existing content in
plan.md(SKILL.md). - Boundary markers: None specified; the skill does not use unique delimiters to separate instructions from architectural data.
- Capability inventory: The skill is designed to perform file system write operations to create and maintain
plan.md. - Sanitization: No sanitization or validation of external input is mentioned before it is formatted into the markdown roadmap.
Audit Metadata