spec-driven-development
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a gated workflow that mandates human validation at every phase (Specify, Plan, Tasks, and Implement). This human-in-the-loop approach serves as a primary security control against unintended or malicious outcomes.
- [CREDENTIALS_UNSAFE]: The skill demonstrates proactive security awareness by including an explicit instruction under 'Boundaries' to 'Never do: Commit secrets'.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests external requirements to generate specifications, the risk is mitigated by the structured 'Success Criteria' reframing and mandatory human review before any code generation or task execution begins.
Audit Metadata