project-management
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- Indirect Prompt Injection (SAFE): The skill handles research literature and notes stored in
.memory/activeContext.md. While this involves external data, the workflow is typical for research tools and lacks evidence of malicious exploitation.\n - Ingestion points: Reads from
projects/{slug}/.memory/activeContext.md.\n - Boundary markers: Not explicitly defined in this skill file, but the data is used for project context persistence.\n
- Capability inventory: Includes project creation, listing, and configuration updates.\n
- Sanitization: Not explicitly mentioned, as the data is primarily intended for the agent's internal memory management.
Audit Metadata