ci-cd-and-automation
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs users to feed raw CI failure output back to the agent for debugging purposes. \n
- Ingestion points: CI failure logs (stdout/stderr) from build, test, and lint runners which can be influenced by project code or external dependencies. \n
- Boundary markers: The suggested prompt template lacks delimiters or specific instructions for the agent to treat the log content as untrusted data. \n
- Capability inventory: The skill demonstrates and utilizes shell commands through tools like
npm,npx,prisma, andvercelacross multiple workflows. \n - Sanitization: No sanitization or filtering of the CI logs is mentioned before providing them to the agent for analysis.
Audit Metadata