Clojure MCP

⚠️ Alpha Software - Work in Progress

This project is in early development and rapidly evolving. While I've found it invaluable for working with Clojure projects and it has significantly improved my development workflow, expect breaking changes, rough edges, and incomplete documentation.

🤝 Help Wanted! If you find this useful, please consider contributing:

• Report bugs and issues you encounter
• Suggest improvements or new features
• Submit pull requests for fixes or enhancements
• Share your configuration patterns and workflows
• Help improve documentation and examples

Your feedback and contributions will help make this tool better for the entire Clojure community!

A Model Context Protocol (MCP) server for Clojure that enables AI assistants (like Claude) to interact directly with a Clojure REPL, providing a collaborative, REPL-driven development workflow between humans and LLMs.

This project implements an MCP server that connects AI models to a Clojure nREPL, enabling a powerful REPL-driven development workflow.

• Clojure REPL Connection
• Clojure Aware editing - Using clj-kondo, parinfer, cljfmt, and clj-rewrite
• Optimized set of tools for Clojure Development
• Emacs edit highlighting - alpha

This approach enables:

• Immediate feedback - Validate code by running it in a stateful REPL, not just statically analyzing it
• Incremental development - Build solutions in small, verified steps
• Human oversight - Keep the programmer in the loop for guidance
• Functional approach - Encourage pure functions that are easier to understand and test

• Clojure (1.11 or later)
• Java (JDK 11 or later)
• Claude Desktop (for the best experience)

Clone this repository to your development machine

or use it as git dep in your deps.edn

In the Clojure project where you want AI assistance, add MCP server configuration to your deps.edn:

Important: Replace /path/to/clojure-mcp with the actual path to your cloned repository.

IMPORTANT NOTE: the mcp server can run in any directory and DOES NOT have to run in your project. The mcp server looks to the nREPL connection for context. The root directory of the project that is running the nREPL server becomes the root directory of all the mcp tool invocations. Currently the nREPL must run on the same machine as the MCP server as there is an assumption of a shared file system between the nREPL server and the MCP server.

Edit your Claude Desktop configuration file:

• Location: ~/Library/Application Support/Claude/claude_desktop_config.json

Replace these paths:

• /path/to/clojure-mcp → Your cloned repository location
• /your/bin/path → Your system's binary path (e.g., /Users/username/.nix-profile/bin)

1. Start nREPL in your target project:

   cd /path/to/your/project clojure -M:nrepl

   You should see: nREPL server started on port 7888...

2. Restart Claude Desktop (required after config changes)

3. Verify connection: In Claude Desktop, click the + button in the chat area. You should see "Add from clojure-mcp" in the menu.

In Claude Desktop click the + tools and optionally add

• resource PROJECT_SUMMARY.md - (have the LLM create this) see below
• resource Clojure Project Info - which introspects the nREPL connected project
• resource LLM_CODE_STYLE.md - Which is your personal coding style instructions (copy the one in this repo)
• prompt clojure_repl_system_prompt - instructions on how to code - cribbed a bunch from Clod Code

Then start the chat.

I would start by stating a problem and then chatting with the LLM to interactively design a solution. You can ask Claude to "propose" a solution to a problem.

Iterate on that a bit then have it either:

A. code and validate the idea in the REPL.

Don't underestimate LLMs abilities to use the REPL! Current LLMs are absolutely fantastic at using the Clojure REPL.

B. ask the LLM to make the changes to the source code and then have it validate the code in the REPL after file editing.

C. ask to run the tests. D. ask to commit the changes.

Make a branch and have the LLM commit often so that it doesn't ruin good work by going in a bad direction.

This project includes a workflow for maintaining an LLM-friendly PROJECT_SUMMARY.md that helps assistants quickly understand the codebase structure.

1. Creating the Summary: To generate or update the PROJECT_SUMMARY.md file, use the MCP prompt in the + > clojure-mcp menu create-project-summary. This prompt will:

   • Analyze the codebase structure
   • Document key files, dependencies, and available tools
   • Generate comprehensive documentation in a format optimized for LLM assistants

2. Using the Summary: When starting a new conversation with an assistant:

   • The "Project Summary" resource automatically loads PROJECT_SUMMARY.md
   • This gives the assistant immediate context about the project structure
   • The assistant can provide more accurate help without lengthy exploration

3. Keeping It Updated: At the end of a productive session where new features or components were added:

   • Invoke the create-project-summary prompt again
   • The system will update the PROJECT_SUMMARY.md with newly added functionality
   • This ensures the summary stays current with ongoing development

This workflow creates a virtuous cycle where each session builds on the accumulated knowledge of previous sessions, making the assistant increasingly effective as your project evolves.

This is NOT required to use the Clojure MCP server.

There are a few MCP tools provided that are agents unto themselves and they need API keys to function.

To use the agent tools, you'll need API keys from one or more of these providers:

• GEMINI_API_KEY - For Google Gemini models

  • Get your API key at: https://makersuite.google.com/app/apikey
  • Used by: dispatch_agent, architect, code_critique

• OPENAI_API_KEY - For GPT models

  • Get your API key at: https://platform.openai.com/api-keys
  • Used by: dispatch_agent, architect, code_critique

• ANTHROPIC_API_KEY - For Claude models

  • Get your API key at: https://console.anthropic.com/
  • Used by: dispatch_agent

Option 1: Export in your shell

Option 2: Add to your shell profile (.bashrc, .zshrc, etc.)

When setting up Claude Desktop, ensure it can access your environment variables by updating your config.

Personally I source them right in bash command:

Note: The agent tools will work with any available API key. You don't need all three - just set up the ones you have access to. The tools will automatically select from available models. For now the ANTHROPIC API is limited to the displatch_agent.

This tool has a learning curve. You may in practice have to remind the LLM to develop in the REPL. You may also have to remind the LLM to use the clojure_edit family of tools which have linters build in to prevent unbalanced parens and the like.

The default tools included in main.clj are organized by category to support different workflows:

Tool Name Description Example Usage

LS	Returns a recursive tree view of files and directories	Exploring project structure
read_file	Smart file reader with pattern-based exploration for Clojure files	Reading files with collapsed view, pattern matching
fs_grep	Fast content search using regular expressions	Finding files containing specific patterns
glob_files	Pattern-based file finding	Finding files by name patterns like *.clj
think	Log thoughts for complex reasoning and brainstorming	Planning approaches, organizing thoughts

Tool Name Description Example Usage

clojure_eval	Evaluates Clojure code in the current namespace	Testing expressions like (+ 1 2)
bash	Execute shell commands on the host system	Running tests, git commands, file operations

Tool Name Description Example Usage

clojure_edit	Structure-aware editing of Clojure forms	Replacing/inserting functions, handling defmethod
clojure_edit_replace_sexp	Modify expressions within functions	Changing specific s-expressions
file_edit	Edit files by replacing text strings	Simple text replacements
file_write	Write complete files with safety checks	Creating new files, overwriting with validation

Tool Name Description Example Usage

dispatch_agent	Launch agents with read-only tools for complex searches	Multi-step file exploration and analysis
architect	Technical planning and implementation guidance	System design, architecture decisions

Tool Name Description Example Usage

code_critique	Interactive code review and improvement suggestions	Iterative code quality improvement

• Collapsed View: Shows only function signatures for large Clojure files
• Pattern Matching: Use name_pattern to find functions by name, content_pattern to search content
• defmethod Support: Handles dispatch values like "area :rectangle" or vector dispatches
• Multi-language: Clojure files get smart features, other files show raw content

• Form-based Operations: Target functions by type and identifier, not text matching
• Multiple Operations: Replace, insert_before, insert_after
• Syntax Validation: Built-in linting prevents unbalanced parentheses
• defmethod Handling: Works with qualified names and dispatch values

• REPL Integration: Executes in the connected nREPL session
• Helper Functions: Built-in namespace and symbol exploration tools
• Multiple Expressions: Evaluates and partitions multiple expressions

• Autonomous Search: Handles complex, multi-step exploration tasks
• Read-only Access: Agents have safe, limited tool access
• Concurrent Execution: Launch multiple agents for parallel processing
• Detailed Results: Returns comprehensive analysis and findings

The Clojure MCP server is designed for easy customization. The codebase separates the core MCP server API from the specific implementation:

• src/clojure_mcp/core.clj - Provides the reusable API for building MCP servers
• src/clojure_mcp/main.clj - Example implementation showing how to use the API

The main.clj file demonstrates the recommended patterns for creating your own customized MCP server:

Resources provide context documents to the AI assistant. Common resources include:

• Project documentation (README, PROJECT_SUMMARY)
• Code style guides
• Project structure information
• Configuration files

Prompts provide pre-configured instructions and workflows for the AI assistant.

Tools are organized by category:

• Read-only: File exploration, project inspection
• Evaluation: Code execution and testing
• Editing: File modification and code generation
• Agents: Advanced AI-powered tools
• Experimental: Cutting-edge features

1. Copy main.clj as a starting point for your custom server
2. Modify the functions to include only the tools/resources/prompts you need
3. Add requires for any additional tools you want to include
4. Update your deps.edn to point to your custom server function

Example deps.edn configuration:

To add tools not included in the default set:

1. Find the tool namespace in src/clojure_mcp/tools/
2. Add the require to your namespace
3. Add the tool function call to your my-tools function

Example:

Choose tools based on your workflow needs:

• For exploration only: Include read-only and evaluation tools
• For active development: Add editing and file manipulation tools
• For AI-assisted workflows: Include agent and experimental tools
• For testing: Include bash tool for running tests

This modular approach lets you create focused, efficient MCP servers tailored to specific development workflows.

Want to create your own MCP tools?

📖 Custom Tools Development Guide

The Clojure MCP server supports project-specific configuration through a .clojure-mcp/config.edn file in your project's root directory. This configuration provides security controls and customization options for the MCP server.

Create a .clojure-mcp/config.edn file in your project root:

Controls which directories the MCP tools can access for security. Paths can be relative (resolved from project root) or absolute.

Boolean flag to enable Emacs integration notifications.

Path Resolution:

• Relative paths (like "src", "../other-project") are resolved relative to your project root
• Absolute paths (like "/home/user/shared") are used as-is
• The project root directory is automatically included in allowed directories

Security:

• Tools validate all file operations against the allowed directories
• Attempts to access files outside allowed directories will fail with an error
• This prevents accidental access to sensitive system files
• the Bash tool doesn't respect these boundaries so be wary

Default Behavior:

• Without a config file, only the project directory and its subdirectories are accessible
• The nREPL working directory is automatically added to allowed directories

Note: Configuration is loaded when the MCP server starts. Restart the server after making configuration changes.

1. Express the problem - Clearly state what you want to solve
2. Develop in the REPL - Work through solutions incrementally
3. Validate step-by-step - Test each expression before moving on
4. Save to files - When the solution is working, save it properly
5. Reload and verify - Make sure the saved code works

• Small steps - Prefer many small, valid steps over a few large steps
• Human guidance - Provide feedback to keep development on track
• Test early - Validate ideas directly in the REPL before committing to them

The core philosophy of this project is that:

1. Tiny steps with rich feedback lead to better quality code
2. REPL-driven development provides the highest quality feedback loop
3. Keeping humans in the loop ensures discernment and maintainable code

GNU Affero General Public License v3.0

Copyright (c) 2025 Bruce Hauman

This program is free software: you can redistribute it and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more details.

You should have received a copy of the GNU Affero General Public License along with this program. If not, see https://www.gnu.org/licenses/.

• ✅ Use freely for personal projects, internal business tools, and development
• ✅ Modify and distribute - improvements and forks are welcome
• ✅ Commercial use - businesses can use this internally without restrictions
• ⚠️ Network copyleft - if you offer this as a service to others, you must open source your entire service stack
• 📤 Share improvements - modifications must be shared under the same license

This license ensures the project remains open source while preventing commercial exploitation without contribution back to the community.