typescript
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill processes untrusted project files and configurations to provide guidance (SKILL.md, Step 1). 1. Ingestion points: Reads local TypeScript source code and tsconfig files. 2. Boundary markers: No explicit markers or 'ignore' instructions for external content are defined in the workflow. 3. Capability inventory: No command execution or network access; capabilities are limited to reading project state and writing markdown reports to /claudedocs/. 4. Sanitization: No sanitization of analyzed code is described. The risk is limited to influencing the agent's analysis or generated report content.
- [Data Exposure] (SAFE): No evidence of hardcoded credentials or access to sensitive system paths like SSH keys was found.
- [Remote Code Execution] (SAFE): No patterns for remote dependency downloads or dynamic code execution were detected.
Audit Metadata