retro
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill provides analytical instructions for improving the development environment based on session history.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from session logs and repository files like package.json and AGENTS.md. While this creates a potential surface for indirect prompt injection, the risk is low as the skill's function is limited to analysis and reporting. Evidence chain: 1. Ingestion points: session logs, package.json, AGENTS.md, CODING_STANDARDS.md. 2. Boundary markers: Absent. 3. Capability inventory: None (the skill body only instructs on analysis and presentation). 4. Sanitization: Absent.
Audit Metadata