ci-test-audit
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data by auditing files in the
test/,.github/, andscripts/ci/directories, creating a surface for instructions embedded in those files to influence the agent. 1. Ingestion points: files intest/,.github/, andscripts/ci/. 2. Boundary markers: none specified for ingested content. 3. Capability inventory: read access for auditing with the constraint to "Propose, do not apply" all changes. 4. Sanitization: none mentioned. - [SAFE]: The skill is a legitimate tool for CI/CD auditing and explicitly requires human review for all actions. It even assists in identifying security risks like untrusted command triggers and anonymous clones.
Audit Metadata