cursor-talk-to-figma-mcp and figma-unified-mcp
These are competitors offering similar MCP-based Figma integration capabilities, with A providing core read/modify functionality while B attempts broader feature coverage (106 tools vs. implicit fewer in A), though A has significantly more adoption and maintenance signals.
About cursor-talk-to-figma-mcp
grab/cursor-talk-to-figma-mcp
TalkToFigma: MCP integration between AI Agent (Cursor, Claude Code) and Figma, allowing Agentic AI to communicate with Figma for reading designs and modifying them programmatically.
Implements a three-tier WebSocket architecture connecting Cursor/Claude to Figma: an MCP server handling tool invocations, a Figma plugin executing design operations, and a WebSocket relay facilitating real-time communication. Provides 40+ MCP tools covering document queries, bulk text replacement, component instance overrides, auto-layout manipulation, annotations, and prototype reaction visualization. Built with TypeScript/Bun and distributed via npm, with native Figma plugin installation via the community marketplace.
About figma-unified-mcp
sso-ss/figma-unified-mcp
Describe it, and it appears in Figma. 106 tools, 14 agent skills. Read, create, edit designs & manage tokens via Claude, Cursor, or any MCP client. Open source.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work