A
licenseA
qualityD
maintenanceA universal MCP server that lets agents discover ComfyUI nodes, build/edit/run API workflows, manage queues, retrieve outputs, and optionally render graphs on a ComfyUI canvas.
23
6
Apache 2.0
comfy-draftsman
Related Servers
Alternatives to comfy-draftsman
No user-submitted related servers found.
Related Servers
- AlicenseBqualityCmaintenanceAn MCP server that bridges AI agents with ComfyUI for automated workflow building, execution, monitoring, and output routing.961MIT
- AlicenseNot gradedqualityDmaintenanceMCP server that enables AI agents to control a local ComfyUI instance for image generation, allowing workflow understanding, parameter modification, execution, and model discovery.16 npm3Apache 2.0
- AlicenseAqualityDmaintenanceMCP server that lets AI agents drive a remote ComfyUI instance end-to-end, including queueing workflows, viewing outputs, downloading models, and authoring graphs via ComfyScript.1612,131 npmMIT
- FlicenseNot gradedqualityDmaintenanceMCP server that connects local ComfyUI to AI agents, enabling natural language control of ComfyUI for creating workflows, generating images, and managing the queue.1-
- AlicenseAqualityBmaintenanceAn MCP server that integrates with a local ComfyUI, allowing assistants to list and run workflows, monitor progress, and view results, as well as read and edit the live workflow canvas via an optional bridge node.6415GPL 3.0
TDQS
A3.9/5.0
Scored across 29 tools
Disambiguation4/5
Descriptions go out of their way to fence off look-alike tools (validate_workflow vs diagnose_workflow vs lint_workflow; resolve_missing_nodes vs port_workflow; check_setup vs get_instance_info), which makes most selections unambiguous. Still, several clusters genuinely overlap — three validation/checking tools, two saved-workflow listings (list_workflows vs find_workflow), and two setup diagnostics — so an agent must read carefully before picking.
Naming Consistency5/5
Every tool is snake_case with a leading verb (inspect_, create_, save_, validate_, list_, search_, get_, run_, manage_, record_), forming a predictable verb_noun convention throughout. Action families are readable at a glance (search_*, list_*, *_workflow).
Tool Count2/5
At 29 tools this is well past the heavy threshold, and a few groupings could be consolidated (three validators, two workflow-listings, two setup checks, search_nodes vs search_node_packs). The domain is broad — authoring, discovery, execution, queueing, assets, learning — so the size is defensible, but it still burdens tool selection.
Completeness4/5
Coverage is unusually deep: full authoring lifecycle (create/import/edit/organize/validate/export/save), node and model discovery, execution with polling, queue management, image upload, and output viewing/relocation. Minor gaps remain, notably no tool to delete or rename workflows already in the browser (save_workflow only renames on collision).
Maintenance
ActivityActive
ResponsivenessNo recent community issues