context-management-context-save
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Data Exposure] (SAFE): The skill defines access to a project root directory to extract metadata and architectural decisions. This behavior is aligned with the stated purpose of context management and is not considered malicious.
- [Indirect Prompt Injection] (SAFE): The skill processes external project files as data ingestion points. While this creates an entry point for untrusted data, it is central to the skill's function. Evidence Chain: 1. Ingestion points: $PROJECT_ROOT files; 2. Boundary markers: Absent; 3. Capability inventory: Local file reading and architectural analysis; 4. Sanitization: The documentation explicitly warns that sensitive information must be excluded.
Audit Metadata