skills/scholarly360/contract-intelligence/intellectual-property-ownership-technology/Gen Agent Trust Hub
intellectual-property-ownership-technology
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No instructions found that attempt to bypass safety filters or override system constraints. The instructions are standard professional guidelines for a legal assistant role.
- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, credential markers, or network operations were identified. The skill operates entirely within the conversation context.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill contains no code or package manifests (e.g., requirements.txt, package.json). It does not perform any remote downloads or script execution.
- [Indirect Prompt Injection] (SAFE): Although the skill is designed to process untrusted user data (contracts), it lacks the exploitable capabilities (such as subprocess execution, file writing, or network access) required to translate an indirect injection into a security breach.
- [Dynamic Execution] (SAFE): No use of eval(), exec(), or runtime compilation was detected. The scripts folder contains only markdown-based guidance.
Audit Metadata