mui-renovate-triage
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill runs local scripts (
collect-prs.mjsandci-logs.mjs) and utilizes the GitHub CLI (gh) for triage operations. Technical review confirms that positional arguments such as repository names and job IDs are validated using regular expressions to prevent command injection. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from PR bodies and CI logs. It mitigates injection risks by instructing the agent to treat this data as information rather than instructions. A capability review shows file write and network operations are restricted to the triage workflow.
- [EXTERNAL_DOWNLOADS]: The skill fetches information from trusted and well-known services (GitHub and CircleCI). These network operations are central to the skill's function and follow standard developer practices.
Audit Metadata