figma-mcp-cached

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow calls figma_prepare_file with a user-supplied figmaUrl, which causes the server to fetch Figma design content (outsider-authored public web content) and then passes that retrieved text/data into the agent via the MCP tool responses (get_figma_data/download_figma_images).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's startup command uses npx to fetch and execute the remote npm package @pactortester/figma-mcp-cached at runtime (npx -y @pactortester/figma-mcp-cached — package page: https://www.npmjs.com/package/@pactortester/figma-mcp-cached), which is a required external dependency that runs remote code.