testing-patterns
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- Indirect Prompt Injection (LOW): The skill's context-detection scripts ingest local project data (e.g., .env, package.json) to customize generated templates. This surface could allow malicious project contents to influence the agent. Ingestion points: scripts/create-integration-test.md, scripts/create-msw-handler.md, scripts/create-page-object.md. Boundary markers: Absent. Capability inventory: File system write operations and test execution command generation. Sanitization: Absent.
- Dynamic Execution (SAFE): The skill utilizes hardcoded shell execution (via ! syntax) for environment discovery. This functionality is used to detect frameworks like pytest or vitest and is a standard part of the skill's primary utility for developer productivity.
Audit Metadata