Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
COMFYUI_PATHYesPath to your local ComfyUI installation.
COMFY_MCP_ALLOWNoAllowlist controlling which workflows/categories the server exposes and runs. Unset exposes only text-to-image and upscale; '*' exposes everything; comma-separated values select categories or workflow ids; empty exposes nothing.

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
list_workflowsA

List all available ComfyUI workflow templates with their categories and descriptions.

get_workflow_paramsA

Get the parameter schema for a specific workflow, including required/optional parameters, types, defaults, and available options.

run_workflowB

Execute a ComfyUI workflow with the specified parameters. Returns generated images.

get_queue_statusB

Get the current ComfyUI queue status (running and pending jobs).

get_job_resultA

Get the result of a completed or running job by its prompt ID. Pass wait_seconds to block until it finishes (repeat the call while status is "running").

cancel_jobB

Cancel a queued or running job.

list_lorasB

Discover available LoRA models categorized by type (qwen, xfs, other).

upload_imageB

Upload an image to ComfyUI for use in image-to-image or controlnet workflows.

download_job_resultsA

Resolve the native ComfyUI output file paths for a completed async job. Use after run_workflow with wait_for_completion=false. Returns absolute paths to the files ComfyUI already wrote in its own output directory — it does not copy anything.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A3.7/5.0

Scored across 9 tools

Disambiguation4/5

Most tools have clearly distinct purposes (list, run, cancel, queue status, upload). The main overlap is between get_job_result and download_job_results, both dealing with job results; descriptions clarify (one returns result data, the other resolves native output paths for async jobs), but an agent could still hesitate over which to call.

Naming Consistency5/5

All nine tools follow a consistent verb_noun snake_case pattern: get_workflow_params, run_workflow, list_workflows, cancel_job, upload_image, etc. No mixing of conventions or vague verbs.

Tool Count5/5

Nine tools is well-scoped for a ComfyUI workflow server, covering discovery, execution, monitoring, cancellation, and asset handling without bloat. Each tool earns its place.

Completeness4/5

The core lifecycle—list workflows, inspect params, run, monitor queue, fetch results, cancel, upload images, locate outputs—is covered. Minor gaps exist: no listing of non-LoRA models (checkpoints/VAEs), no queue-clearing or job-history listing, but these are workaroundable.

Maintenance

ActivityMaintained
ResponsivenessNo issues