gimp-mcp

gimp-mcp

Enables AI agents to perform GIMP-style image operations such as open, resize, crop, flip, rotate, blur, desaturate, text overlay, export, and batch processing via MCP tools, supporting both mock (Pillow) and live GIMP backends.

Category
Visit Server

README

gimp-mcp

Python 3.11+ Version MCP License: MIT MergeOS

gimp-mcp is an MCP server so AI agents can drive GIMP-style image operations: open/create images, resize, crop, flip, rotate, blur, desaturate, text overlay, export, and batch resize.

Mode Backend When
mock (default) Pillow CI, offline, no GIMP install
live gimp-console batch + Pillow assist Machine with GIMP 2.10 / 3.x

Product: mergeos-bounties/gimp-mcp · Funded: prj_0525


Install (one command)

Grok — recommended

pip install "git+https://github.com/mergeos-bounties/gimp-mcp.git" && grok plugin install mergeos-bounties/gimp-mcp --trust

This installs the Python CLI (gimp-mcp) and the Grok plugin (skill + MCP server from .mcp.json).

Check:

gimp-mcp version
gimp-mcp doctor
gimp-mcp demo
grok plugin list
grok mcp list

Local clone:

git clone https://github.com/mergeos-bounties/gimp-mcp.git
cd gimp-mcp
pip install -e ".[dev]"
grok plugin install . --trust

Other agents (stdio MCP)

After pip install "git+https://github.com/mergeos-bounties/gimp-mcp.git", point any MCP host at:

Field Value
command gimp-mcp
args ["serve"]
env GIMP_MCP_MODE=mock

Claude Desktop — merge examples/claude_desktop_config.json into Claude MCP config.

Cursor — merge examples/cursor_mcp.json.

Grok config.toml (manual, without plugin):

[mcp_servers.gimp_mcp]
command = "gimp-mcp"
args = ["serve"]
env = { GIMP_MCP_MODE = "mock" }
enabled = true

One-liner via Grok CLI:

pip install "git+https://github.com/mergeos-bounties/gimp-mcp.git"
grok mcp add gimp-mcp -- gimp-mcp serve

Supported AI agents / hosts

Host Support Install
Grok (CLI / TUI / Build) Yes grok plugin install mergeos-bounties/gimp-mcp --trust then pip install "git+https://github.com/mergeos-bounties/gimp-mcp.git"
Claude Desktop Yes Copy examples/claude_desktop_config.json into Claude MCP settings
Cursor Yes Merge examples/cursor_mcp.json into Cursor MCP config
Claude Code Yes stdio MCP: same command/args as Claude Desktop / Grok
VS Code (MCP / Continue / Cline) Yes Generic stdio server config pointing at gimp-mcp serve
Windsurf / Cascade Yes stdio MCP entry with gimp-mcp + serve
Codex CLI Yes (stdio) Register MCP server command gimp-mcp serve in Codex MCP settings
ChatGPT Desktop Partial Only if host supports custom MCP stdio servers
Gemini CLI Partial Only if MCP stdio plugins are enabled

All packages speak MCP over stdio (gimp-mcp serve). Default mode is mock (offline, no simulator/terminal/GIMP required).


Highlights

Capability Description
MCP tools gimp_doctor, gimp_open, gimp_resize, gimp_export, …
CLI gimp-mcp demo, doctor, live-smoke, serve, call
Offline demo Always works with Pillow mock
Live GIMP 3 python-fu-eval + --quit (~2s scale on Windows)
Live GIMP 2 Script-Fu batch
Grok plugin skills/gimp-mcp + .mcp.json + .grok-plugin/plugin.json

Quick start (developers)

cd mcp\GIMP-mcp
pip install -e ".[dev]"
gimp-mcp version
gimp-mcp doctor
gimp-mcp demo
pytest -q

Live GIMP (optional)

# Install GIMP 3.x, then:
$env:GIMP_MCP_MODE = "live"
# optional override:
# $env:GIMP_MCP_BIN = "$env:LOCALAPPDATA\Programs\GIMP 3\bin\gimp-console.exe"
gimp-mcp doctor
gimp-mcp live-smoke

MCP host config

Plugin ships .mcp.json. Manual examples: examples/claude_desktop_config.json, examples/cursor_mcp.json.

{
  "mcpServers": {
    "gimp-mcp": {
      "command": "gimp-mcp",
      "args": ["serve"],
      "env": { "GIMP_MCP_MODE": "live" }
    }
  }
}

CLI / tools

Command Purpose
gimp-mcp version Package + detected console
gimp-mcp doctor Backend health
gimp-mcp demo Mock end-to-end smoke
gimp-mcp live-smoke Live console smoke (skips if missing)
gimp-mcp tools list MCP tool names
gimp-mcp call … One-shot tool (key=value)
gimp-mcp process <image> Full real-photo pipeline → multiple exports
gimp-mcp serve Stdio MCP server

Real photos

$env:GIMP_MCP_MODE = "live"
gimp-mcp process "C:\path\to\photo.jpg" --out-dir ".\examples\real-run" --watermark "MergeOS"

Writes *_processed.png, *_gray.png, *_crop_flip_*.png, and batch_thumbs/ (GIMP 3 scale via python-fu-eval).

MCP tools: gimp_mode, gimp_doctor, gimp_seed_demo, gimp_list_images, gimp_close, gimp_new_image, gimp_open, gimp_info, gimp_resize, gimp_thumbnail, gimp_crop, gimp_flip, gimp_rotate, gimp_blur, gimp_sharpen, gimp_desaturate, gimp_invert, gimp_brightness, gimp_contrast, gimp_auto_orient, gimp_text_overlay, gimp_pipeline, gimp_export, gimp_batch_resize.

Env:

Variable Meaning
GIMP_MCP_MODE mock (default) or live
GIMP_MCP_BIN Path to gimp-console
GIMP_MCP_WORKSPACE Working directory for temp images
GIMP_MCP_TIMEOUT Batch timeout seconds

Development

pip install -e ".[dev]"
ruff check src tests
pytest -q

Plugin layout:

.grok-plugin/plugin.json   # Grok plugin manifest
.mcp.json                  # MCP stdio server for Grok
skills/gimp-mcp/           # Skill (auto-loaded after plugin install)
.grok/skills/gimp-mcp/     # Same skill for in-repo discovery

MergeOS bounties

Follow mergeos-bounties + star mergeos and mergeos-contracts. Claim issues → PR to master → 25–200 MRG.

See docs/BOUNTY.md.


License

MIT — see LICENSE.

Live test note (maintainer)

Verified on Windows with GIMP 3.2.4 at
%LOCALAPPDATA%\Programs\GIMP 3\bin\gimp-console.exe

  • Live resize uses python-fu-eval + --quit (not Script-Fu hang)
  • gimp-mcp live-smoke completes in a few seconds
  • Filters may still use Pillow assist by design

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured