An MCP-first control plane with cryptographic model identity, explicit approvals, and durable run evidence for the ComfyUI setup you already use.
local-gpu-imagegen is an MCP project maintained in the chevalgrand520/local-gpu-imagegen repository. The maintainer describes its purpose as “An MCP-first control plane with cryptographic model identity, explicit approvals, and durable run evidence for the ComfyUI setup you already use”. Read the project documentation for the complete scope.
The project provides a local client configuration using python. Follow the repository’s build and installation steps before copying it, and replace example paths with the paths on your own computer.
Project documentation · Repository · Discovery listing. This profile records the public documentation checked on 2026-10-08.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| local_gpu_discover_models | Plan or execute bounded API/filesystem inventory without loading model weights. |
| local_gpu_inspect_workflow | Inspect one ordinary ComfyUI API txt2img workflow and return diagnostic or registerable hashes and bindings. |
| local_gpu_register_workflow | Recheck an exact proposal and store its immutable registered workflow copy after later digest-bound confirmation. |
| local_gpu_set_model_trust | Inspect a reviewed component bundle without mutation, or approve/revoke one exact identity in user-local state after confirmation. |
| local_gpu_recommend_models | Return one deterministic route and at most two explained alternatives. |
| local_gpu_list_profiles | List registered use-case profiles and the current backend capabilities. |
| local_gpu_start_run | Persist a confirmed intent, profile, constraints, backend choice, and round budget. |
| local_gpu_get_run | Read the durable manifest and its recoverable_next_actions. |
| local_gpu_branch_run | Create an immutable child run from one reviewed parent round and preserve/change contract. |
| local_gpu_prepare_mask | Prepare a user or rectangle/polygon mask and return a bounded JPEG overlay. |
| local_gpu_confirm_mask | Confirm an unchanged prepared mask after explicit user approval. |
| local_gpu_generate_round | Generate one root or fixed-mode child round and optionally return a bounded JPEG preview. |
| local_gpu_record_review | Store rubric scores, required structured visual checks, hard failures, constraint and preservation results, critique, and next action. |
| local_gpu_finalize_run | Verify the image-bound user confirmation and publish the nominated eligible round as the final local PNG. |
| local_gpu_cleanup_run | Remove intermediates or the entire confirmed run directory. |
| local_gpu_imagegen_check | |
| local_gpu_generate_image |
{
"mcpServers": {
"local-gpu-imagegen": {
"command": "python",
"args": [
"<project-root>\\scripts\\mcp_server.py"
]
}
}
}From the project documentation. Replace placeholders and example paths before connecting.
The ML hub in your context window — search models, datasets, papers and run Spaces from the official server.
Thousands of open models on tap — run image, video and audio generation through Replicate's hosted API.
This community Workers MCP demo exposes image generation, page screenshots and image extraction through a local client adapter and Cloudflare backend.
Integrates Flux's image generation capabilities into AI coding assistants, enabling image generation, manipulation, and control directly within IDEs.
Generates images from text prompts using the Together AI platform through a Model Context Protocol (MCP) server.
Edit images with simple text commands using AI.