git-work-recap
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes PR descriptions and issue tracker comments. The risk is minimized by the skill's read-only nature and its focus on content authored by the user. 1. Ingestion points: PR titles, bodies, and tracker activity (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Read-only data retrieval via gh CLI and git (SKILL.md). 4. Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill uses the gh CLI and git commands to fetch necessary metadata. These are standard developer tools used for their intended purpose within a read-only context. Evidence: gh pr list, gh pr view, gh api, git remote get-url (SKILL.md).
Audit Metadata