running-in-ci

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill extensively utilizes system commands for CI automation tasks.
  • Uses the GitHub CLI (gh) to view PRs, issues, diffs, and check statuses.
  • Employs git for branch management and pushing code updates to repositories.
  • Utilizes jq to parse and extract information from JSONL session logs.
  • Uses bash for logic such as polling loops to monitor CI progress.
  • [EXTERNAL_DOWNLOADS]: The skill downloads artifacts from external sources during the troubleshooting process.
  • Downloads session log artifacts from GitHub Actions workflow runs using gh run download in SKILL.md.
  • These downloads originate from GitHub, which is a trusted service.
  • [PROMPT_INJECTION]: The skill processes data from untrusted sources, making it a surface for indirect prompt injection.
  • Ingestion points: Pull request descriptions, issue bodies, triggering comments, PR diffs, and session log artifacts.
  • Boundary markers: None explicitly defined for isolating untrusted text within the logic.
  • Capability inventory: The skill can modify the repository state via git push, create or update pull requests, and post comments or replies to GitHub threads.
  • Sanitization: The skill relies on agent instructions for "Grounded Analysis" rather than programmatic sanitization or filtering of the input content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 03:45 AM