triage
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to verify pull requests by checking out the code and running 'relevant tests or commands'. This involves executing code provided by external, untrusted contributors. The skill attempts to mitigate this by instructing the agent to 'Review code and scripts before running contributor tests in isolation with no privileged secrets.'
- [COMMAND_EXECUTION]: The workflow involves querying issue trackers via CLI tools (e.g.,
gh issue list) and executing repository-specific build or test commands as part of the triage verification process. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from issue bodies, comments, and PR diffs. It includes defensive instructions for the agent to treat these inputs as 'untrusted context' and ensures that triage notes or agent briefs do not override repository rules or user authorization.
Audit Metadata