📌Overview

Purpose: The clojure-sdk is designed to facilitate the creation of Model Context Protocol (MCP) servers through a robust and user-friendly interface.

Overview: This framework provides developers with the necessary tools to quickly set up MCP servers, showcasing various functionalities through built-in examples like calculators and data visualizers. The SDK is built on top of lsp4clj, ensuring efficient handling of JSON-RPC based communications while simplifying development tasks.

Key Features:

Calculator Server: Offers a variety of basic arithmetic operations, allowing users to perform tasks such as addition, subtraction, or calculating averages directly through the MCP interface.
Vega-lite Server: Enables the generation of sophisticated visualizations using Vega-lite. It includes features like data visualization and chart generation, requiring a pre-installed vl-convert CLI for operation.

io.modelcontext/clojure-sdk

A clojure-sdk for creating Model Context Protocol servers!

Usage

The examples contain full working code for defining MCP servers including:

calculator_server
vegalite_server
code_analysis_server

The examples is a deps-new app project. Instructions for compiling and running the example servers are available in the examples README.

The dependencies for clojure-sdk are:

{io.modelcontextprotocol/mcp-clojure-sdk
 {:git/url "https://github.com/unravel-team/mcp-clojure-sdk.git"
  :git/sha "039cf220ac6bb3858f71e823016035e257a5380d"}}

Examples

Building the Examples Jar

$ make clean && make examples-jar

The examples jar contains the following servers:

Calculator: calculator_server
Vega-lite: vegalite_server
Code Analysis: code_analysis_server

Calculator: `calculator_server`

Provides basic arithmetic tools: add, subtract, multiply, divide, power, square-root, average, factorial.

Example commands:

What's the average of [1, 2, 3, 4, 5]?
What's the factorial of 15?
What's 2 to the power of 1000?
What's the square-root of 64?

Before running the calculator MCP server:

Use the full-path to the examples JAR on your system.

In Claude Desktop

{
  "calculator": {
    "command": "java",
    "args": [
      "-Dclojure.tools.logging.factory=clojure.tools.logging.impl/log4j2-factory",
      "-Dorg.eclipse.jetty.util.log.class=org.eclipse.jetty.util.log.Slf4jLog",
      "-Dlog4j2.contextSelector=org.apache.logging.log4j.core.async.AsyncLoggerContextSelector",
      "-Dlog4j2.configurationFile=log4j2-mcp.xml",
      "-Dbabashka.json.provider=metosin/jsonista",
      "-Dlogging.level=INFO",
      "-cp",
      "/Users/vedang/mcp-clojure-sdk/examples/target/io.modelcontextprotocol.clojure-sdk/examples-1.2.0.jar",
      "calculator_server"
    ]
  }
}

In MCP Inspector

npx @modelcontextprotocol/inspector java -Dclojure.tools.logging.factory=clojure.tools.logging.impl/log4j2-factory -Dorg.eclipse.jetty.util.log.class=org.eclipse.jetty.util.log.Slf4jLog -Dlog4j2.contextSelector=org.apache.logging.log4j.core.async.AsyncLoggerContextSelector -Dlog4j2.configurationFile=log4j2-mcp.xml -Dbabashka.json.provider=metosin/jsonista -Dlogging.level=INFO -cp examples/target/io.modelcontextprotocol.clojure-sdk/examples-1.2.0.jar calculator_server

Vega-lite: `vegalite_server`

Provides tools for generating Vega-lite charts: save-data, visualize-data.

Pre-requisite: Needs vl-convert CLI to be installed.

Example data sample:

[
    { "year": 2011, "value": 14.6, "growth_type": "Market Cap Growth" },
    { "year": 2011, "value": 11.4, "growth_type": "Revenue Growth" },
    { "year": 2011, "value": 26.6, "growth_type": "Net Income Growth" },
    { "year": 2012, "value": 40.1, "growth_type": "Market Cap Growth" },
    { "year": 2012, "value": 42.7, "growth_type": "Revenue Growth" },
    { "year": 2012, "value": 36.9, "growth_type": "Net Income Growth" },
    { "year": 2013, "value": 16.9, "growth_type": "Market Cap Growth" },
    { "year": 2013, "value": 14.6, "growth_type": "Revenue Growth" },
    { "year": 2013, "value": 15.3, "growth_type": "Net Income Growth" },
    { "year": 2014, "value": 9.6, "growth_type": "Market Cap Growth" },
    { "year": 2014, "value": 7.9, "growth_type": "Revenue Growth" },
    { "year": 2014, "value": 10.9, "growth_type": "Net Income Growth" },
    { "year": 2015, "value": 5.8, "growth_type": "Market Cap Growth" },
    { "year": 2015, "value": 6.7, "growth_type": "Revenue Growth" },
    { "year": 2015, "value": 6.2, "growth_type": "Net Income Growth" },
    { "year": 2016, "value": -12.4, "growth_type": "Market Cap Growth" },
    { "year": 2016, "value": -3.9, "growth_type": "Revenue Growth" },
    { "year": 2016, "value": -32.2, "growth_type": "Net Income Growth" },
    { "year": 2017, "value": 25.3, "growth_type": "Market Cap Growth" },
    { "year": 2017, "value": 5.9, "growth_type": "Revenue Growth" },
    { "year": 2017, "value": 43.9, "growth_type": "Net Income Growth" }
]

Visualize this data using Vega-lite.

Before running the vegalite MCP server

Replace the full-path to the examples JAR with the correct path on your system.
Specify the full-path to vl-convert on your system.

In Claude Desktop

{
  "vegalite": {
    "command": "java",
    "args": [
      "-Dclojure.tools.logging.factory=clojure.tools.logging.impl/log4j2-factory",
      "-Dorg.eclipse.jetty.util.log.class=org.eclipse.jetty.util.log.Slf4jLog",
      "-Dlog4j2.contextSelector=org.apache.logging.log4j.core.async.AsyncLoggerContextSelector",
      "-Dlog4j2.configurationFile=log4j2-mcp.xml",
      "-Dbabashka.json.provider=metosin/jsonista",
      "-Dlogging.level=INFO",
      "-Dmcp.vegalite.vl_convert_executable=/Users/vedang/.cargo/bin/vl-convert",
      "-cp",
      "/Users/vedang/mcp-clojure-sdk/examples/target/io.modelcontextprotocol.clojure-sdk/examples-1.2.0.jar",
      "vegalite_server"
    ]
  }
}

In MCP Inspector

npx @modelcontextprotocol/inspector java -Dclojure.tools.logging.factory=clojure.tools.logging.impl/log4j2-factory -Dorg.eclipse.jetty.util.log.class=org.eclipse.jetty.util.log.Slf4jLog -Dlog4j2.contextSelector=org.apache.logging.log4j.core.async.AsyncLoggerContextSelector -Dlog4j2.configurationFile=log4j2-mcp.xml -Dbabashka.json.provider=metosin/jsonista -Dlogging.level=INFO -Dmcp.vegalite.vl_convert_executable=/Users/vedang/.cargo/bin/vl-convert -cp examples/target/io.modelcontextprotocol.clojure-sdk/examples-1.2.0.jar vegalite_server

Code Analysis: `code_analysis_server`

Provides prompts: analyse-code and poem-about-code.

You can try the prompts in Claude Desktop or Inspector as a demonstration of exposing powerful prompts.

Before running the code-analysis MCP server

Replace the full-path to the examples JAR with the correct path on your system.

In Claude Desktop

{
  "code-analysis": {
    "command": "java",
    "args": [
      "-Dclojure.tools.logging.factory=clojure.tools.logging.impl/log4j2-factory",
      "-Dorg.eclipse.jetty.util.log.class=org.eclipse.jetty.util.log.Slf4jLog",
      "-Dlog4j2.contextSelector=org.apache.logging.log4j.core.async.AsyncLoggerContextSelector",
      "-Dlog4j2.configurationFile=log4j2-mcp.xml",
      "-Dbabashka.json.provider=metosin/jsonista",
      "-Dlogging.level=INFO",
      "-cp",
      "/Users/vedang/mcp-clojure-sdk/examples/target/io.modelcontextprotocol.clojure-sdk/examples-1.2.0.jar",
      "code_analysis_server"
    ]
  }
}

In MCP Inspector

npx @modelcontextprotocol/inspector java -Dclojure.tools.logging.factory=clojure.tools.logging.impl/log4j2-factory -Dorg.eclipse.jetty.util.log.class=org.eclipse.jetty.util.log.Slf4jLog -Dlog4j2.contextSelector=org.apache.logging.log4j.core.async.AsyncLoggerContextSelector -Dlog4j2.configurationFile=log4j2-mcp.xml -Dbabashka.json.provider=metosin/jsonista -Dlogging.level=INFO -cp examples/target/io.modelcontextprotocol.clojure-sdk/examples-1.2.0.jar code_analysis_server

Pending Work

You can see the list of pending changes in the todo.org file.

Development of the SDK

The clojure-sdk is a standard deps-new project. Expected commands:

Run tests:

$ make test
# or
clojure -T:build test

Run CI and build JAR:

$ make build
# or
clojure -T:build ci

This produces an updated pom.xml inside target/classes/META-INF and the JAR in target. Update version and SCM tag information in build.clj.

Install locally:

$ make install
# or
clojure -T:build install

Deploy to Clojars (requires CLOJARS_USERNAME and CLOJARS_PASSWORD environment variables and building first):

$ make deploy
# or
clojure -T:build deploy

The library will be deployed to io.modelcontext/clojure-sdk on clojars.org.

Inspiration

This SDK is built on top of lsp4clj, which handles the edge-cases of JSON-RPC servers. It provides a smart and easy-to-modify foundation.

License

Distributed under the MIT License

mcp-clojure-sdk

📌Overview

io.modelcontext/clojure-sdk

Usage

Examples

Building the Examples Jar

Calculator: calculator_server

Before running the calculator MCP server:

In Claude Desktop

In MCP Inspector

Vega-lite: vegalite_server

Before running the vegalite MCP server

In Claude Desktop

In MCP Inspector

Code Analysis: code_analysis_server

Before running the code-analysis MCP server

In Claude Desktop

In MCP Inspector

Pending Work

Development of the SDK

Inspiration

License

Calculator: `calculator_server`

Vega-lite: `vegalite_server`

Code Analysis: `code_analysis_server`