project-status
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is explicitly designed to be read-only, instructing the agent not to modify code, documentation, or git/GitHub state.
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) to retrieve information about pull requests and issues relevant to the project status. - [COMMAND_EXECUTION]: To verify implementation claims, the skill may execute local diagnostic tools or tests (e.g.,
npm test). The instructions specifically limit this to established, fast diagnostics with understood side effects. - [PROMPT_INJECTION]: The skill processes potentially untrusted data from repository files like roadmaps and TODO lists. It mitigates the risk of indirect prompt injection by requiring the agent to 'triangulate' and verify these claims against independent evidence (such as commit history or implementation code) before including them in the briefing.
Audit Metadata