figma-console-mcp-skills

Warn

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation instructions recommend cloning a repository from an unverified GitHub user ('PercentProduction/figma-console-mcp-skills-347').
  • [REMOTE_CODE_EXECUTION]: Following the installation process leads to running 'npm install' and 'npm start' on code downloaded from an unverified external repository.
  • [COMMAND_EXECUTION]: The skill references the use of shell scripts ('.sh') for certain operations, allowing for arbitrary command execution on the host system.
  • [PERSISTENCE]: The configuration section recommends adding sensitive environment variables to the user's shell profile ('~/.zshrc'). While intended for configuration, modifying shell profiles is a method used to maintain persistence.
  • [PROMPT_INJECTION]: The skill processes untrusted data from external sources, specifically Figma comments and design data, and possesses capabilities to modify files and perform network operations.
  • Ingestion points: Content is fetched from the Figma REST API and the Figma Plugin API as described in 'SKILL.md'.
  • Boundary markers: No delimiters or safety warnings are present to separate untrusted data from instructions.
  • Capability inventory: Capabilities include the 'use_figma' tool for Figma interaction, network access via 'fetch', and local process execution.
  • Sanitization: No logic for sanitizing external content is identified in the provided code snippets.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 15, 2026, 12:04 PM
Security Audit — agent-trust-hub — figma-console-mcp-skills