Hatchling MCP Server

What is Hatchling?

Hatchling is an interactive CLI-based chat application that integrates local Large Language Models (LLMs) through Ollama and OpenAI with the Model Context Protocol (MCP) for tool calling capabilities. It serves as the frontend for using all MCP servers in the Hatch ecosystem.

How to use Hatchling?

Installation and running are guided through Docker setup and configuration steps provided in the documentation. Users can extend Hatchling with custom MCP tools by creating Hatch packages using the hatch:create command and defining tools with the @hatch_mcp.tool() decorator, then adding the package with hatch:pkg:add and connecting via mcp:server:connect.

Key features of Hatchling

• CLI-based chat interface with syntax highlighting and command auto-completion
• Integrates Ollama and OpenAI LLMs with MCP tool calling
• Tool execution wrapping to support longer tool calling chains
• Extensible via custom Hatch packages for new MCP servers
• Displays token usage, MCP server lifecycle, and tool chaining information

Use cases of Hatchling

• Building custom AI assistants that leverage MCP tools for specialized tasks
• Integrating local LLMs with biological data analysis and modeling software
• Managing collections of MCP servers with environment and package management
• Developing and testing new MCP tools using the Python SDK decorator patterns