ComfyUI-MCP lets developers and AI artists control ComfyUI's visual workflows with natural language, enabling real-time content generation and graph editing locally.
229 upvotes
Tracked by Kanon since Aug 2, 2026
🤖 Agent teardown · skill
Job to be doneUsers want to generate images, video, and audio, and edit ComfyUI workflows in real time using natural language commands.
AutonomyL2 · tool-calling(evidence: "edits your live ComfyUI graph in natural language…")
Who it is forDevelopers, AI artists, and users of local AI workflows
Prerequisitesopen source · self-hostable · bring your own API key
Integrationsmcp · comfyui · ollama · discord · runpod
PricingFree
Traction · why it is rising229 GitHub upvotes indicate early adoption and validation within the AI creator community.
Why it matters
It bridges natural language and ComfyUI’s visual interface, serving as a foundational local control layer for AI agents with potential as a reusable agent desktop shell.
Signal source: GitHub
Visit official site →
On mobile tap Share for WeChat / RED (Xiaohongshu) / X; on desktop use Copy text and paste into the app.