gh-fix-ci
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and analyzes GitHub Actions logs, which are external inputs that could be manipulated by an attacker to include malicious instructions.\n
- Ingestion points: The script
scripts/inspect_pr_checks.pyfetches logs from the GitHub API and GitHub CLI (gh run view --log).\n - Capability inventory: The skill is intended to draft and implement fixes for CI failures, which involves the ability to write to the local filesystem and modify code.\n
- Sanitization: Log content is extracted as failure snippets and presented directly to the agent without sanitization or filtering of prompt injection patterns.\n
- Boundary markers: No explicit delimiters or boundary markers are used to separate the untrusted log data from the agent's primary instructions.
Audit Metadata