contract-review
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional prompts designed to guide an AI agent through a multi-step contract review process. It uses established legal frameworks (CUAD, LegalBench) to identify risks in documents.
- [EXTERNAL_DOWNLOADS]: The documentation references an external tool,
legal-redline-tools, which can be installed via pip from the author's GitHub repository. This is presented as an optional companion utility for formatting the skill's output into documents. As this resource belongs to the skill's vendor ('evolsb'), it is considered a legitimate dependency for the described workflow. - [DATA_EXPOSURE]: The skill processes user-uploaded contracts for analysis but does not contain any instructions or tools for external network communication, file exfiltration, or credential harvesting.
- [SAFE]: Analysis of the instructions and examples shows no evidence of prompt injection, obfuscation, or persistence mechanisms. The skill operates within the typical constraints of an AI assistant.
Audit Metadata