gh-fix-ci
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
subprocess.runto execute localgitandgh(GitHub CLI) commands to retrieve repository information, PR status, and CI logs. The implementation uses argument lists rather than shell strings, which is a security best practice to prevent command injection.\n- [INDIRECT_PROMPT_INJECTION]: The skill fetches and analyzes external GitHub Actions logs. These logs are treated as untrusted data that could contain malicious instructions designed to influence the agent's behavior during the failure summarization or fix-planning phase.\n - Ingestion points: Logs are fetched via
gh run view --logandgh apiinscripts/inspect_pr_checks.py.\n - Boundary markers: The skill does not explicitly use delimiters or specific ignore instructions warnings when presenting log snippets to the agent.\n
- Capability inventory: The agent can execute local commands, write files, and implement code fixes.\n
- Sanitization: The script extracts snippets based on failure keywords but does not sanitize the log content for prompt injection patterns.
Audit Metadata