Vibe Coding Discover

MCP

comfyui-mcp

View on GitHub

Local-first, agent-native control plane for ComfyUI — MCP server + sidebar agent that generates images, video & audio, authors and runs workflows, and edits your live graph in natural language on ANY LLM (Claude, ChatGPT, Gemini, offline Ollama, or any hosted model). 178 tools, 36 AI skills, 55 installer packs. Local, LAN, VPS, or Comfy Cloud.

★ 778133 forksTypeScriptMITartokun

An MCP server and sidebar agent that lets LLMs operate ComfyUI: generate media, build and edit workflows, manage models and custom nodes, and troubleshoot runs. Supports local, remote, and cloud setups, including offline models via Ollama.

Use Cases

Generate images, video, and audio in ComfyUI through natural-language requestsCreate, edit, and run ComfyUI workflowsManage models and custom nodesDiagnose failed workflows and suggest fixesCompare and optimize workflow settingsRun ComfyUI with local models through OllamaInstall model-specific workflows and custom-node packs

Built With

Language
TypeScript
Frameworks
Model Context Protocol · ComfyUI · Ollama

Tags

MCP server · AI agent · ComfyUI · image generation · video generation · audio generation · workflow automation · local-first · agent skills · Claude Code · Ollama · model management

comfyui-mcp — Vibe Coding Discover