pipe
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes instructions for the agent to process untrusted data from external sources, specifically GitHub repository metadata such as Pull Request titles, issue bodies, and commit messages, to automate workflow generation and auditing.
- Ingestion points: Data enters the context via GitHub event payloads (e.g.,
${{ github.event.pull_request.title }}) during PR automation and workflow design tasks. - Boundary markers: The skill mandates the use of explicit boundary logic, instructing the agent to 'pass untrusted values via env: and quote them' rather than using direct interpolation.
- Capability inventory: The agent's capabilities are restricted to generating and auditing GitHub Actions YAML configuration files. There are no included scripts or tools that allow for arbitrary shell execution or system-level privilege escalation.
- Sanitization: The skill provides clear sanitization requirements, such as avoiding the use of
evaland requiring the use ofprintfor environment variable mapping for all user-controlled strings.
Audit Metadata