Project Atlantis

Name: Project Atlantis
Author: ProjectAtlantis-dev

@ProjectAtlantis-dev

リポジトリ

Project Atlantis について

A configurable MCP client/server

基本情報

カテゴリ

その他

ランタイム

python

トランスポート

stdio

公開者

ProjectAtlantis-dev

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "atlantis-mcp-server": {
      "command": "python",
      "args": [
        "server.py",
        "\\"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Project Atlantis?

Project Atlantis is a Python MCP remote server that lets users install custom functions as tools and manage third‑party MCP servers on the fly. It connects to an experimental cloud server at projectatlantis.ai for tool sharing and authentication, and is designed for developers exploring MCP capabilities.

How to use Project Atlantis?

Install Python 3.12 and Node, then configure the runServer script with your email and API key, and sign in at projectatlantis.ai. Run the Python server process, which will auto-connect to the cloud.

Key features of Project Atlantis

Dynamic functions as hot‑reloaded Python tools
Dynamic MCP server management via JSON config
Cloud‑connected for auth and tool sharing
Supports multiple remote instances
Local MCP client for use with Claude or Cursor

Use cases of Project Atlantis

Experimenting with MCP protocol and remote server concepts
Building custom tools with hot‑reloaded Python functions
Managing and sharing third‑party MCP servers
Running multiple remote instances across machines

FAQ from Project Atlantis

What is the difference between a remote and an MCP server?

The remote is the Python MCP host process, while MCP servers are third‑party tools that can be dynamically installed via JSON config files.

What are the runtime requirements?

Python 3.12 (most stable) and Node for the MCP client, plus uv/uvx and npx for MCP tooling.

How does authentication work?

Authentication is currently handled through the Atlantis cloud at projectatlantis.ai using your email and an API key (default foobar).

Where do custom functions live?

Custom functions are stored in the dynamic_functions/ folder and are automatically loaded on start with hot‑reload support.

What transports does the server support?

The remote uses a WebSocket cloud connection, and a local npx‑based MCP client that does not require the cloud (but only exposes tools on the local box).