brainstorming
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE] (SAFE): The skill consists exclusively of markdown instructions and a Graphviz diagram. It contains no executable scripts, shell commands, or external dependencies.
- [Indirect Prompt Injection] (LOW): The skill defines a process where the agent reads project files and documentation to gain context, which is an attack surface for indirect prompt injection. 1. Ingestion points: The agent reads 'files, docs, recent commits' during the 'Explore project context' phase. 2. Boundary markers: No explicit boundary markers or 'ignore' instructions for external content are defined. 3. Capability inventory: The agent is authorized to write design documents to the '.context/plans/' directory and perform git commits. 4. Sanitization: No sanitization process is described for content read from the project environment.
Audit Metadata