fix-ci
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external content from GitHub Actions logs and external check links to diagnose and resolve errors, creating an indirect prompt injection risk if log outputs contain malicious instructions.
- Ingestion points: Step 2 in
SKILL.mddirects the agent to inspect failed jobs via GitHub Actions logs or external links. - Boundary markers: Absent. There are no delimiters or instructions specifying that the agent should ignore instructions embedded inside the log text or external link content.
- Capability inventory: The skill possesses the capability to modify the codebase to apply fixes and push code changes back to the repository (
SKILL.md). - Sanitization: Absent. No validation or filtering is defined to sanitize the log inputs before processing them.
Audit Metadata