juliadoc
MCP server for efficiently retrieving Julia docstrings and source code
claude mcp add --transport stdio jonathanfischer97-juliadoc-mcp npx -y @jonathanfischer97/server-juliadoc \ --env JULIA_PROJECT="/path/to/your/julia/project"
How to use
The Juliadoc MCP server provides tools to access Julia documentation and source code through Claude Desktop. It exposes two main commands: get-doc, which fetches documentation for a Julia package, module, type, function, or method (for example Base.sort or AbstractArray), and get-source, which retrieves the corresponding Julia source code. The server uses a lightweight, built-in caching mechanism with a five-minute TTL to speed up repeated requests. To use it, configure your Claude Desktop client to point at the juliadoc MCP server and invoke the available tools with the required path parameter. If you have a Julia project you want to reference, set JULIA_PROJECT in the environment so the server loads your local environment; otherwise it will fall back to the global environment on your system.
How to install
Prerequisites:
- Node.js 16 or higher
- npm (comes with Node.js)
- Julia 1.9 or higher installed and accessible in PATH
- Claude Desktop installed and configured to connect to MCP servers
Installation steps:
-
Clone the MCP server repository (or use the npm package if published): git clone https://github.com/jonathanfischer97/juliadoc-mcp.git cd juliadoc-mcp
-
Install dependencies: npm install
-
Build the server (if the project requires a build step): npm run build
-
Start the server locally for testing: npm start
-
In Claude Desktop, add the MCP server configuration as shown in the README ( mcpServers.juliadoc with command npx and the package name). Ensure Julia is installed and accessible in PATH. If you want Claude Desktop to reference a specific Julia project, set JULIA_PROJECT in the environment as shown in the example configuration.
Additional notes
Environment and configuration tips:
- If JULIA_PROJECT is set, the server will use that Julia project environment; otherwise it will use the default global environment. Ensure the specified path is accessible to the process.
- The server relies on the system's Julia executable being discoverable in PATH. If you encounter a Julia not found error, verify your PATH includes the Julia bin directory.
- The caching TTL is 5 minutes by default; if you need fresher results for rapidly changing documentation, you can adjust your workflow to force re-fetches when appropriate.
- When using npx with a scoped package, ensure you have network access and the correct npm registry permissions.
- For debugging, run the server locally and check logs for errors related to Julia environment loading or network requests from Claude Desktop.
Related MCP Servers
frontmcp
TypeScript-first framework for the Model Context Protocol (MCP). You write clean, typed code; FrontMCP handles the protocol, transport, DI, session/auth, and execution flow.
shinzo-ts
TypeScript SDK for MCP server observability, built on OpenTelemetry. Gain insight into agent usage patterns, contextualize tool calls, and analyze server performance across platforms. Integrate with any OpenTelemetry ingest service including the Shinzo platform.
openai -agent-dotnet
Sample to create an AI Agent using OpenAI models with any MCP server running on Azure Container Apps
mcp-bun
Bun Javascript Runtime MCP Server for AI Agents
lichess
A Model Context Protocol (MCP) server that enables Claude Desktop to interact with Lichess chess platform through natural language. Play games, analyze positions, manage your account, and participate in tournaments—all by simply talking to Claude.
mcp -starter
MCP (model context protocol) starter server. For UI libraries, APIs, open-source projects and more