nic01asfr/qgisremotemcp

Эта страница переведена автоматически. Технические детали смотрите в оригинале. Открыть оригинал

11 stars 3 forks Качество 74 Тренд 22MIT

QGIS Desktop as an MCP Server — Full GUI via noVNC, PyQGIS scripting, 1000+ Processing algorithms, smart data pipeline with 30+ French national datasets.

Обзор

Give any AI assistant a complete GIS workstation — data loading, spatial analysis, cartography, and multi-format export — all running in a live QGIS Desktop instance. is an MCP server that exposes a full QGIS Desktop through Docker. The AI loads data, runs analysis, produces maps — while users interact with the same instance in their browser via noVNC. : Production-ready for single-user local deployments. Multi-user mode (per-user isolated containers with GPU passthrough) is functional but in beta. Developed at Cerema Méditerranée — published to share the approach and invite contributions. Everything runs inside Docker — the AI calls MCP tools, QGIS does the work, the user sees results live in their browser. No local QGIS installation needed. The REST API, the MCP server and noVNC check every non-loopback call (src/workspace_auth.py). /health and calls from 127.0.0.1 stay open.

README

Показан фрагмент оригинального README. Открыть оригинал.

QgisRemoteMCP — QGIS Desktop as an MCP Server

Give any AI assistant a complete GIS workstation — data loading, spatial analysis, cartography, and multi-format export — all running in a live QGIS Desktop instance.

QgisRemoteMCP is an MCP server that exposes a full QGIS Desktop through Docker. The AI loads data, runs analysis, produces maps — while users interact with the same instance in their browser via noVNC.

Project status: Production-ready for single-user local deployments. Multi-user mode (per-user isolated containers with GPU passthrough) is functional but in beta. Developed at Cerema Méditerranée — published to share the approach and invite contributions.

What it does

You say to the AI	What happens
“Analyse le risque inondation à Nîmes”	Loads flood zones + buildings, computes exposure, exports interactive HTML with water height slider
“Carte de densité du bâti à Montpellier”	Downloads 10,000 buildings as local GPKG, creates hex grid, graduated symbology, PDF A3 export
“Prépare un relevé terrain pour Sète”	Loads data, styles layers, exports QField-ready ZIP with editable Observations layer (camera, dropdowns)
“Analyse l’évolution foncière sur la côte”	DVF transactions 2020-2024, coastal distance bands, temporal web map with animated playback

Everything runs inside Docker — the AI calls MCP tools, QGIS does the work, the user sees results live in their browser. No local QGIS installation needed.

Quick start

1. Run

git clone https://github.com/nic01asFr/QgisRemoteMCP.git
cd QgisRemoteMCP
cp .env.example .env
docker compose up -d --build

2. Verify

Service	URL
MCP Server	http://localhost:8100/mcp
noVNC (QGIS in browser)	http://localhost:6080
REST API	http://localhost:8081
MJPEG stream	http://localhost:8082/stream

curl http://localhost:8100/health
# → {"status":"ok","bridge":true,"server":"QgisRemoteMCP","multi_user":false}

Access control

The REST API, the MCP server and noVNC check every non-loopback call (src/workspace_auth.py). /health and calls from 127.0.0.1 stay open.

Variable	Effect
`WORKSPACE_AUTH_MODE`	`permissive` (default: unauthenticated calls are logged, then served), `enforce` (refused with 401), `off`

Открыть полный README на GitHub

Установка

This server does not publish a one-line install command.

Open the repository installation guide Новичок в MCP? Начните с руководства на mcpskill.org →

Конфигурация

{
  "mcpServers": {
    "qgis": {
      "type": "http",
      "url": "http://localhost:8100/mcp"
    }
  }
}