brainstorm
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to explore the local codebase, including project structure, documentation, and recent commits, to inform the brainstorming process. This creates an attack surface where malicious content embedded in the repository files could attempt to influence the agent's behavior during the brainstorming session.
- Ingestion points: Project structure, documentation files, and git commit history via codebase exploration.
- Boundary markers: The skill uses a
<HARD-GATE>tag to define implementation constraints, but does not specify delimiters for the ingested repository data. - Capability inventory: The agent uses file system and git tools to read project data.
- Sanitization: No explicit sanitization or filtering of codebase content is mentioned.
Audit Metadata