ProjectAtlantis-dev/atlantis-mcp-server
If you are the rightful owner of atlantis-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
Project Atlantis is an experimental platform that leverages the Model Context Protocol (MCP) to provide agentic capabilities to bots, allowing for local control and dynamic function management.
Project Atlantis is a platform designed to explore the potential of the Model Context Protocol (MCP) by providing a Python-based remote server that allows users to install and manage dynamic functions and third-party MCP tools. Unlike traditional cloud architectures, MCP enables more local control, which can be both a benefit and a challenge due to the increased setup complexity. The project includes a Python MCP host, referred to as a 'remote', which can be configured to connect to the Atlantis cloud server. Users can create custom functions that are automatically reloaded upon modification, and manage third-party MCP servers through JSON configuration files. The platform is still in development, with ongoing work on MCP authentication and security.
Features
- Dynamic Functions: Allows users to create and maintain custom functions-as-tools, which are automatically reloaded when modified.
- Dynamic MCP Servers: Enables installation and management of third-party MCP server tools through JSON configuration files.
- Local and Cloud Integration: Supports both local execution and remote control via the Atlantis cloud, providing flexibility in deployment.
- Automatic Reloading: Functions and dependencies are automatically reloaded upon modification, ensuring up-to-date execution.
