agent-testing-patterns
Pass
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation resource, providing legitimate Python code examples for implementing test suites for AI agents.
- [INDIRECT_PROMPT_INJECTION]: The skill describes test workflows for agents that process user-provided prompts, creating a potential surface for indirect injection in the described systems.
- Ingestion points:
real_agent.run(prompt)in E2E test examples (SKILL.md) - Boundary markers: Not defined in the educational code snippets
- Capability inventory: Demonstrates tool use (
FileReadTool,SearchTool) and syntax validation - Sanitization: Not demonstrated in the simplified test patterns
- [DYNAMIC_EXECUTION]: The end-to-end testing section includes a snippet that uses Python's
compile()function. This is used specifically to validate the syntax of code generated by an agent during a test, rather than to execute untrusted code at runtime.
Audit Metadata