generate-project-plan
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill performs codebase 'grounding' by reading up to 20 files (e.g., README.md, ARCHITECTURE.md, OWNERS). This is an intended feature for context gathering and is limited to the repository root with a hard cap on the number of files read. This behavior is considered safe as it is bounded and restricted to the user's project context.
- [DYNAMIC_EXECUTION]: The skill generates and executes JavaScript snippets via the use_figma MCP tool to interact with the Figma API. This is the primary mechanism for board creation and follows provided templates for various FigJam blocks (sections, tables, diagrams). The execution is limited to the Figma plugin environment and uses pre-defined patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from PRDs and codebase files. It mitigates the risk of indirect prompt injection by using an interactive flow where the user must confirm proposed sections and content before any writes occur, and by providing screenshots for review at each major step.
- [EXTERNAL_DOWNLOADS]: The skill loads external 'Foundation skills' (figma-use, figma-use-figjam, figma-generate-diagram) which are official tools from the vendor (Figma). These references are documented neutrally as they originate from a trusted vendor.
Audit Metadata