implementation-contract-review-harness
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected. The skill is composed of instructional markdown files and configuration for evaluation tests.- [PROMPT_INJECTION]: The skill includes evaluation cases (evals/evals.json) specifically designed to test the agent's resistance to user pressure. For example, it validates that the agent refuses to approve low-quality contracts even when prompted by the user to 'Approve it anyway' or 'Rewrite everything from scratch'.- [NO_CODE]: The skill contains no executable scripts (Python, JavaScript, Shell, etc.) and relies entirely on natural language instructions for the agent to perform its review task.- [DATA_EXFILTRATION]: No network access tools or commands are listed in the skill's requirements or logic, and no external URLs are referenced for data transmission.- [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic execution patterns present in the analyzed files.
Audit Metadata