brainstorm
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill is designed to read local implementation files and project context as part of its brainstorming workflow. This access is limited to gathering information relevant to the user's idea and follows a structured context loading protocol.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests content from external files (Step 1) and has the capability to write files (Step 6). However, the risk is minimal as the skill's primary function is to generate documentation, and it requires iterative user validation via the Question Tool throughout the workflow.
- Ingestion points: Reads relevant implementation files and project context in Step 1.
- Boundary markers: None explicitly defined to separate project data from instructions.
- Capability inventory: File system write access to the
docs/brainstorms/directory. - Sanitization: No specific sanitization or filtering of external content is mentioned before it is processed into the design brief.
Audit Metadata