unit-test-generator
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to analyze and extract the contract of user-supplied code (inputs, outputs, dependencies, and assumptions) to generate tests. This process creates a surface for indirect prompt injection where malicious instructions could be embedded in the target code being analyzed.
- Ingestion points: The skill operates on 'target code' provided by the user during the discovery phase (SKILL.md).
- Boundary markers: The instructions use markdown headers and structured taxonomy tables to separate the analysis process from the code, but they do not define technical delimiters for the untrusted input code.
- Capability inventory: The agent following these instructions is expected to generate and potentially write unit test files, which involves file system operations.
- Sanitization: The skill does not provide explicit guidance on sanitizing or escaping the target code content before processing it for test generation.
Audit Metadata