Comfyui MCP logo

Comfyui MCP

未认领

作者:artokun

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.

agent-skillsai-agentclaude-codeclaude-plugincomfyuicomfyui-extensioncomfyui-mcpcomfyui-mcp-serverfluximage-generationlocal-firstmcpmcp-servermodel-context-protocolofflineollamaself-hostedstable-diffusionvideo-generationwan

01 / 选择客户端

02 / 连接前的准备

认证方式尚未指定。连接前请查看项目说明。

安装项目所需的运行环境,并确保客户端能够使用它。

项目说明

输入项目说明中的变量名,用逗号分隔。只输入名称,不要输入密钥值。

03 / 添加配置

来源:根据公开安装说明生成

claude mcp add --transport stdio 'comfyui-mcp' -- 'npx' '-y' 'comfyui-mcp@latest' '--tunnel'

替换占位符后在终端中运行。

04 / 在客户端中确认

打开客户端的 MCP 设置,确认服务器已连接并列出工具。复制配置不代表连接成功。

更多AI & ML服务器

浏览完整目录

未认领列表

这个 MCP 服务器是你的吗?

此列表根据公开信息自动生成。认领后,你可以编辑页面、设置兼容性并解锁增长工具。全程不超两分钟。

认领此服务器

安全概况

已认领和已认证的服务器每周接受一次静态扫描,显示代码能触及的范围(外部服务、环境变量、shell 命令、代理配置目录)以及带有已知公告的依赖。认领此列表即可获得。 安全概况的工作原理

12 个工具中显示 12 个

文档中列出的工具 (12)

内容来自项目文档。服务器握手不会验证每个工具的说明或行为。

clear_vram

Free GPU VRAM by unloading cached models — calls ComfyUI's /free endpoint, reports before/after stats

comfy_cli

Search actual loaded node classes through official comfy nodes search (action:"searchnodes")

create_workflow

One action-parameterized tool for authoring: create (from templates: txt2img, img2img, upscale, inpaint, controlnet, ipadapter, acestep15, stableaudio3), modify (operations: setinput, addnode, removenode, connect, insertbetween), validate (dry-run — missing nodes, broken connections, invalid output

download_model

Find and fetch model weights, and track the transfers. action: download (from a URL, into the correct ComfyUI subdirectory), status, cancel, search (HuggingFace), searchcivitai, searchcreators, downloadcivitai, resolvemissing

get_system_stats

Get system info — GPU, VRAM, Python version, OS

get_workflow

One action-parameterized tool for READING a saved workflow file: list (the user library, subfolders included), get (one workflow's JSON by filename), analyze (a structured summary instead of raw JSON), query (filter/traverse/aggregate a big graph without dumping it), strip (de-virtualize any workflo

list_local_models

What is installed, and where ComfyUI looks. action: list (installed models by type: checkpoints, loras, vae, upscalemodels, controlnet, embeddings, clip, unet, diffusionmodels, textencoders), remove (deletes a model file), embeddings, listpaths, addpath, removepath (the last three view/edit the extr

queue

One action-parameterized tool for the execution queue: list (running + pending), status (one job by prompt ID), getworkflow (a pending job's full payload), move/edit (requeue a pending job front/back, patched or replaced, with a new prompt ID), cancel (interrupt the running job — escalates interrupt

save_workflow

One action-parameterized tool for WRITING to the library: save (store a workflow — overwrites a same-filename file), lock (record a provenance lock: SHA-256 per model, git commit per node pack), verifylock (report drift against that lock)

search_custom_nodes

Search the ComfyUI Registry for node packs by keyword (action: "search"), or get one pack's full details (action: "details")

upload_image

Copy a local image into ComfyUI's input/ directory for img2img, inpaint, or ControlNet

visualize_workflow

One action-parameterized tool for rendering and converting a workflow you pass in: render (Mermaid flowchart, nodes grouped by category), renderhierarchical (the same graph sectioned — overview, one section in detail, a listing, or an AI-oriented summary), mermaid (a Mermaid diagram back to executab

工具变更历史

比较相同配置的完整检查。只列出工具,未调用工具。此处不测量输入结构的变化。

尚无完整工具检查。