context-engineering-project
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security threats detected. The skill is purely instructional, focusing on repository context organization and refining local rules files (e.g., AGENTS.md, CLAUDE.md).
- [SAFE]: Indirect Prompt Injection Surface Analysis: 1. Ingestion points: The skill instructs the agent to read repository source code, external specifications, and configuration files. 2. Boundary markers: Includes specific instructions to treat instruction-like text in external materials as data rather than commands to follow. 3. Capability inventory: Rely on the agent's inherent capabilities to read/write files and execute development commands within a repository. 4. Sanitization: Emphasizes distinguishing trust levels of sources and verifying outputs against existing patterns.
Audit Metadata