figma-create-design-system-rules
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFE
Full Analysis
- Project Configuration and Rule Generation: The skill is designed to create and modify project-level configuration files including CLAUDE.md, AGENTS.md, and Cursor rules. This involves generating instructional content based on existing project patterns to guide future agent actions.
- Indirect Data Ingestion Surface: The skill ingests data from two main sources: the local codebase and design data from the Figma MCP server. 1. Ingestion points: Codebase file paths (identified in SKILL.md Step 2) and Figma node data/metadata (identified in SKILL.md Step 1 and 3). 2. Boundary markers: The skill provides templates for structured output but does not specify delimiters for incoming untrusted data during processing. 3. Capability inventory: Access to read repository files, execute Figma MCP tools, and write to project configuration files. 4. Sanitization: No specific sanitization methods for ingested content are detailed. While this creates a data processing surface, it is standard for the skill's intended purpose of codebase analysis.
- Local Helper Scripts: The inclusion of scripts/check_agents_md.sh allows the skill to verify the existence of rule files. The script is a standard bash utility that performs a read-only check and uses best-practice error handling.
- Figma MCP Integration: The skill integrates with the Figma Model Context Protocol (MCP) server to fetch design context and screenshots, facilitating the translation of design to code according to project-specific conventions.
Audit Metadata