uw-unity-test-runner
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external Gherkin documents (e.g.,
docs/GDD.md) to generate executable C# code, creating an indirect prompt injection surface. - Ingestion points: Reads test scenarios from
docs/GDD.mdas specified in SKILL.md. - Boundary markers: No specific delimiters or instructions are provided to the agent to help it distinguish between scenario data and potentially malicious embedded instructions.
- Capability inventory: The skill allows the agent to create C# test files and trigger their execution via the
run_teststool. - Sanitization: There is no evidence of input validation or content sanitization for the source GDD scenarios.
Audit Metadata