create-verification-skill
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes the target repository's codebase to generate verification instructions, which involves processing untrusted local project data as an ingestion point. 1. Ingestion points: repository interview phase (Step 1) and feature map seeding (Step 3). 2. Boundary markers: Absent. 3. Capability inventory: writing files to the project directory, building binaries, and starting/tearing down local app instances. 4. Sanitization: Absent.
- [DYNAMIC_EXECUTION]: The skill generates a new project-local skill and executes its verification procedure to ensure the harness works correctly. This behavior is the primary intended function of the tool and follows simple template-based generation.
Audit Metadata