1. Conduid
  2. AI
  3. Arion
MCP server · AI

Arion

Arion is the first agentic AI desktop app for geospatial analysis, developed by GeoRetina Inc.

Unclaimed GPL-3.0 last commit 6 months ago ai
72Good

Scored a month ago · breakdown

About Arion

Arion is an MCP server published by GeoRetina in the AI category: arion is the first agentic AI desktop app for geospatial analysis, developed by GeoRetina Inc.

The repository has 86 stars and 29 forks, with the last commit 6 months ago. Six months or more without a commit doesn't mean the server is broken, but check the open issues before depending on it in production.

Install

From source
git clone https://github.com/GeoRetina/Arion

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Arion

Powered by Claude · Grounded in docs

I know everything about Arion. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

Static checks derived from the repository 18 hours ago. Heuristics on the source, not an audit.

  • ✓README presentThe repository ships documentation.
  • ✓License declaredUsable under a stated license.
  • ✓Tests presentA test suite exists in the repository.
  • ✓Dependencies pinnedA lockfile pins dependency versions.
  • !Scoped permissionsDoesn't declare a permission scope. Assume it can do anything its process can.

1 check has not been run yet.

README

Arion: Cross‑Platform Desktop App for Agentic Geospatial AI Analysis

Arion is a cross-platform desktop application designed for advanced geospatial analysis and agentic workflows. Built with Electron, React (TypeScript), and Vite, Arion runs natively on Windows, macOS, and Linux, empowering users to leverage local and cloud-based Large Language Models (LLMs), integrate custom Model Context Protocol (MCP) servers, and utilize a plugin system for extended capabilities.

🎥 Demo Video

Core Features

  • Interactive Chat Interface: Communicate with LLMs for geospatial queries, analysis, and task automation. Supports various providers (OpenAI, Google Gemini, Azure, Anthropic, Ollama) via the Vercel AI SDK.
  • Dynamic Map Visualization: Render and interact with geospatial data using MapLibre GL.
  • LLM Tool Integration:
    • Built-in tools for map manipulation (add features, buffers, set view), UI control, and knowledge base queries.
    • Support for user-defined MCP servers, allowing agents to access external Python/TypeScript tools and data sources.
  • Knowledge Base Integration: Ingest documents (PDF, DOCX, TXT) into a local vector store (PGlite with pgvector) for Retrieval Augmented Generation (RAG).
  • Local LLM Support: Configure and use local LLMs via Ollama.
  • Agentic Workflows (Planned): Future support for running Python (LangGraph/CrewAI) and TypeScript agents in isolated processes, orchestrated by the application.
  • Plugin System (Planned): Extend Arion's functionality with custom plugins for data connectors, visualization providers, MCP tools, and agent providers.
  • SQLite/SpatiaLite Backend: Manages application settings, chat history, and plugin configurations. SpatiaLite for advanced geospatial data operations (integration in progress).

Technology Stack

  • Desktop Framework: Electron
  • Frontend: React, TypeScript, Vite, Tailwind CSS
  • Backend (Main Process): Node.js, TypeScript
  • Mapping: MapLibre GL
  • AI/Chat: Vercel AI SDK, LangChain (for future agent runtimes)
  • LLM Tools: Model Context Protocol (MCP) - @modelcontextprotocol/sdk
  • Local Vector Store (Knowledge Base): PGlite with pgvector extension
  • Database (Application Data): SQLite (using better-sqlite3)
  • Build & Packaging: Electron Vite, Electron Builder
  • Linting & Formatting: ESLint, Prettier

Project Structure

  • src/main/: Electron Main process code (Node.js environment).
  • src/renderer/src/: Electron Renderer process code (React UI).
  • src/preload/: Electron Preload script for secure IPC.
  • src/shared/: Code shared between Main and Renderer (e.g., IPC types).
  • plugins/ (Planned): Top-level directory for user-installed plugins.
  • python-agents/ (Planned): Directory for Python-based agent runtimes.

Recommended IDE Setup

Project Setup

Prerequisites

  • Node.js 22 or higher
  • npm 8 or higher

Install Dependencies

npm install

Development

Run the application in development mode with hot reloading:

npm run dev

Build for Production

Arion uses native Node modules (better-sqlite3, keytar) that require platform-specific build tools.

Windows

Requires Visual Studio Build Tools (C++ workload) or windows-build-tools:

npm run build:win

Produces an NSIS installer (arion-<version>-setup.exe) for x64.

To build without code signing:

npm run build:win:unsigned

macOS

Requires Xcode Command Line Tools:

xcode-select --install
npm run build:mac

Produces a .dmg disk image.

Linux

Requires build-essential, python3, and packaging dependencies:

# Ubuntu / Debian
sudo apt install -y build-essential python3 libsecret-1-dev

npm run build:linux

Produces AppImage, Snap, and .deb packages.

Build without Packaging

To compile without creating distributable installers (useful for testing):

npm run build:unpack

Rebuild Native Modules

After updating dependencies, rebuild native modules for your Electron version:

npm run rebuild

Raster GDAL Runtime

  • Windows builds include bundled GDAL binaries.
  • macOS and Linux use bundled GDAL first, then automatically fall back to system GDAL binaries from PATH if bundled binaries are not present.
  • See resources/gdal/README.md for supported layouts and ARION_GDAL_* overrides.

Install GDAL (Quick)

Use these commands to install GDAL quickly on each OS:

# macOS (Homebrew)
brew install gdal

# Ubuntu / Debian
sudo apt update && sudo apt install -y gdal-bin libgdal-dev

# Fedora / RHEL
sudo dnf install -y gdal gdal-devel

# Arch Linux
sudo pacman -S --needed gdal

Windows:

  • Arion does not use system GDAL fallback on Windows, and shipped Windows app builds already include GDAL.
  • For local development or preparing bundled binaries, install GDAL with Conda:
    • conda install -c conda-forge gdal

Verify install:

gdalinfo --version

Contributing

We welcome contributions from the community! Please see our Contributing Guide for details on:

  • 🚀 Getting started with development
  • 📋 Types of contributions we're looking for
  • 🛠️ Development guidelines and coding standards
  • 📝 Licensing terms for contributors

Quick start: Fork the repository, make your changes, and submit a pull request.

For questions: support@georetina.com

License

Arion is free software released under the GNU General Public License v3.0 or later.

  • You may use, modify, and redistribute the project under the GPL terms (including commercial use).
  • Distributions and derivative works must provide source code and remain licensed under the GPL.
  • See the LICENSE file for the complete text and guidance on applying the GPL to your changes.

README fetched from the repository 18 hours ago. The original is authoritative.

Questions

About Arion

How do I install Arion?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Arion safe to use with an AI agent?

Its trust score is 72 out of 100 (good). It passes 4 of the 5 static security checks run so far; the failures are listed above. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Arion still maintained?

The last commit was 6 months ago and 4 open issues. That's long enough that you should check whether the maintainer is responding to issues before depending on it.