context-engineering
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious patterns or behaviors were detected. The skill serves as a purely instructional framework for project management and coding conventions.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface area for indirect prompt injection by instructing the agent to read external files (e.g., CLAUDE.md, AGENTS.md, or issue comments). However, the skill provides explicit security guidelines to mitigate this risk.
- Ingestion points: Reads repository configuration files, source code, and user-provided content.
- Boundary markers: Explicitly defines trust levels for different input types and mandates that user content be treated as untrusted.
- Capability inventory: The skill provides no additional scripts or subprocess tools; it relies on the agent's existing environment.
- Sanitization: Instructs the agent to validate untrusted content at boundaries and verify configuration files before reliance.
Audit Metadata