parallel-ci-triage
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) to list workflow runs and retrieve logs for debugging failed CI jobs. - [COMMAND_EXECUTION]: Directs the agent to execute standard development commands such as
npm run lint,pytest, andpnpm testto verify fixes locally before pushing code. - [DATA_EXPOSURE]: Includes a explicit security recommendation for the user to redact sensitive information (secrets) from logs before providing them to the agent.
- [INDIRECT_PROMPT_INJECTION]: The workflow involves processing logs from external CI runs which could theoretically contain malicious content.
- Ingestion points: Log excerpts are ingested from the
ghcommand or manual copy-pasting of GitHub Actions logs. - Boundary markers: None explicitly defined in the subagent instruction template.
- Capability inventory: Subagents are granted capabilities to edit source code and execute local shell commands for verification.
- Sanitization: No automated sanitization; the skill relies on manual redaction and user oversight.
Audit Metadata