A
licenseNot graded
qualityB
maintenanceMCP server that wraps ComfyUI for SDXL image generation. It exposes tools for generating images, listing models, and checking ComfyUI health, with presets and GPU resource coordination.
1
MIT
comfyui-mcp
Related Servers
Alternatives to comfyui-mcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityCmaintenanceEnables MCP clients to connect to and interact with ComfyUI for workflow execution and image generation.1Apache 2.0
- AlicenseAqualityBmaintenanceWraps ComfyUI to generate images, video, and upscales from text prompts via MCP tools, providing a curated workflow interface without the node editor.84MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI image generation from text prompts by wrapping local ComfyUI as a Streamable HTTP MCP server, supporting synchronous generation, queue management, and history tracking.1MIT
- FlicenseNot gradedqualityBmaintenanceMCP server that exposes ComfyUI image and 3D generation as a single tool with smart prompt classification. It enables AI assistants to generate assets directly by selecting and running the appropriate ComfyUI workflow.-
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients to generate and edit images using local Stable Diffusion models in ComfyUI, with tools for text-to-image, img2img, ControlNet, UI frames, upscaling, and background removal—no cloud API or API key required.MIT
TDQS
A3.9/5.0
Scored across 41 tools
Disambiguation2/5
Several tools have overlapping boundaries: comfy_cli duplicates queue, get_history, upload_image, and download_model via its actions; generate_image overlaps with create_workflow + enqueue_workflow; install_comfyui and install_custom_node both manage custom nodes. Despite the descriptions, an agent faces ambiguous choices for common tasks like listing jobs or running a workflow, which will lead to misselection.
Naming Consistency4/5
Most tools follow a predictable pattern: action-oriented tools use verb_noun (get_workflow, create_workflow, download_model) while domain-oriented tools use noun names (queue, apps, batch, kitchen). The split is deliberate and readable, with only minor deviations like 'bisect' and 'calculate' that are single verbs rather than verb_noun.
Tool Count2/5
41 tools is excessive even for a broad ComfyUI automation domain (workflow authoring, execution, models, training, cloud pods, etc.). Although some consolidation already exists via action-driven mega-tools, the surface remains heavy and could be reduced by merging related single tools.
Completeness4/5
The surface covers nearly all headless ComfyUI workflows: generation, workflow CRUD, queue management, history, models, custom node management, training, cloud pods, and batch runs. The main gap is the absence of live-canvas panel tools (panel_*), which are repeatedly referenced but not included, preventing direct canvas inspection or manipulation from this server alone.
Maintenance
ActivityInactive
ResponsivenessNo issues