test-before-deploy
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: This skill possesses an attack surface for indirect prompt injection due to its core function of processing untrusted project data.
- Ingestion points: Ingests data from the repository, requirements, CI configuration, incidents, and analytics as defined in the Phase 1 workflow of SKILL.md.
- Boundary markers: The skill defines a source-fidelity contract using specific tags ([PDF], [I], [PROJECT], [EXTERNAL]) to label information sources, which provides a logical separation but not a technical boundary for untrusted data.
- Capability inventory: The skill allows the agent to inspect project directories, analyze CI/CD configurations, and generate test implementation code based on the project's existing framework.
- Sanitization: The instructions do not specify any sanitization, escaping, or validation of the untrusted project data before it is incorporated into the agent's reasoning process.
Audit Metadata