brainstorming
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFE
Full Analysis
- [Indirect Prompt Injection] (SAFE): The skill processes untrusted local project data to inform the brainstorming process. 1. Ingestion points: Project files, documentation, and recent git commits referenced in the 'Understanding the idea' section. 2. Boundary markers: Absent; there are no explicit instructions to ignore embedded prompts in the analyzed project files. 3. Capability inventory: File-writing to the 'docs/plans/' directory and git operations (commit, worktree). 4. Sanitization: Absent; project content is used directly for context. Risk is considered safe as it is limited to the local development environment and serves the primary purpose of the skill.\n- [Command Execution] (SAFE): The skill utilizes git commands for committing design documents and managing isolated workspaces via worktrees. These actions are standard for a technical design assistant and do not pose a security risk in this context.
Audit Metadata