pathfinding
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill consists exclusively of Markdown documentation and contains no executable scripts or binaries.
- [SAFE] (SAFE): No malicious behaviors such as exfiltration, obfuscation, or privilege escalation were found. The skill only directs the AI's reasoning processes.
- [Indirect Prompt Injection] (SAFE): The skill reads project files like CLAUDE.md, creating a surface for indirect injection. This is assessed as safe because the skill lacks dangerous capabilities. 1. Ingestion points: User input and project metadata files. 2. Boundary markers: Absent. 3. Capability inventory: Uses internal planning tools; no OS command or network access. 4. Sanitization: Absent, but risk is minimal due to lack of hazardous tools.
Audit Metadata