Limps
limps your Local Intelligent MCP Planning Server across AI assistants. No subscriptions, no cloud—run it locally. Version control your planning docs in git. No more context drift—one shared source of truth across Claude, Cursor, Copilot, and any MCP
What is Limps?
Limps is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to limps your local intelligent mcp planning server across ai assistants. no subscriptions, no cloud—run it locally. version control your planning docs in git. no more context drift—one shared source of ...
limps your Local Intelligent MCP Planning Server across AI assistants. No subscriptions, no cloud—run it locally. Version control your planning docs in git. No more context drift—one shared source of truth across Claude, Cursor, Copilot, and any MCP
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- limps your Local Intelligent MCP Planning Server across AI a
Use Cases
Maintainer
Works with
Installation
NPM
npx -y @sudosandwich/limpsManual Installation
npx -y @sudosandwich/limpsConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Limps
LIMPS (Local Intelligent MCP Planning Server) is a TypeScript MCP server that provides a shared, version-controlled planning and documentation layer across all your AI tools — Claude, Cursor, Copilot, and any other MCP-compatible agent. It stores plans, tasks, and documentation in local Markdown files tracked by git, exposing tools for document management, task planning, knowledge graph queries, and health checking. With no cloud dependency and no subscription required, LIMPS eliminates context drift when working across multiple AI assistants on the same project.
Prerequisites
- Node.js 18 or higher installed
- npm 7+ installed alongside Node.js
- Git installed (LIMPS encourages version-controlling your planning docs)
- Claude Desktop, Cursor, Copilot Chat, or another MCP-compatible AI client
- A directory to store your planning documents (can be an existing project directory)
Install LIMPS globally via npm
Install the @sudosandwich/limps package globally so the limps CLI is available system-wide.
npm install -g @sudosandwich/limps
# Verify installation
limps --versionInitialize a LIMPS workspace
Navigate to the directory where you want to store planning documents and run limps init. This creates a .limps/ directory with the default configuration.
cd ~/Documents/my-project-plans
limps initPrint the MCP client configuration
LIMPS can generate the exact JSON config snippet for your MCP client. Run this command and copy the output into your client's config file.
limps config print --client claude-code
# Or for Claude Desktop:
limps config print --client claude-desktopStart the LIMPS server
Start the LIMPS daemon manually to verify it runs without errors before adding it to your MCP client config.
limps server start
limps server statusAdd LIMPS to your Claude Desktop configuration
Edit your claude_desktop_config.json to add the LIMPS MCP server entry. Use npx with the -y flag so no global install is required, or reference the global install path.
Restart Claude Desktop and create your first plan
After restarting, ask Claude to create a plan or document. Verify that a Markdown file appears in your planning directory and can be tracked with git.
# After testing with Claude, version control your plans
cd ~/Documents/my-project-plans
git init
git add .
git commit -m 'initial planning docs'Limps Examples
Client configuration
Claude Desktop configuration using npx to run LIMPS without a global install. The workspace path tells LIMPS where to store planning documents.
{
"mcpServers": {
"limps": {
"command": "npx",
"args": ["-y", "@sudosandwich/limps"],
"env": {
"MCP_PLANNING_CONFIG": "/absolute/path/to/your/plans-directory/.limps/config.json"
}
}
}
}Prompts to try
These queries use the document management, planning, and knowledge graph tools exposed by LIMPS.
- "Create a plan for the user authentication refactor with tasks broken into subtasks."
- "What is the current status of the Q3 migration plan?"
- "Search my planning docs for anything related to database schema changes."
- "What is the next highest priority task in the current active plan?"
- "Check for any stale or drifted documents that haven't been updated in over 2 weeks."Troubleshooting Limps
limps command not found after npm install -g
The global npm bin directory may not be on your PATH. Run 'npm config get prefix' to find it, then add <prefix>/bin to your PATH in ~/.zshrc or ~/.bashrc. Alternatively use npx @sudosandwich/limps instead of the global command.
LIMPS server fails to start with 'config not found' error
Run 'limps init' in the directory you want to use as your planning workspace before starting the server. This creates the required .limps/config.json file. If using MCP_PLANNING_CONFIG, ensure the path is absolute and the file exists.
Context from LIMPS is not shared between Claude Desktop and Cursor
Both clients must point to the same .limps/config.json file via the MCP_PLANNING_CONFIG environment variable. Verify both config files reference the same absolute path. Run 'limps config print' from the planning directory to get the correct config snippet for each client.
Frequently Asked Questions about Limps
What is Limps?
Limps is a Model Context Protocol (MCP) server that limps your local intelligent mcp planning server across ai assistants. no subscriptions, no cloud—run it locally. version control your planning docs in git. no more context drift—one shared source of truth across claude, cursor, copilot, and any mcp It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Limps?
Install via npm with the command: npx -y @sudosandwich/limps. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with Limps?
Limps works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Limps free to use?
Yes, Limps is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Limps Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to Limps? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Limps in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Limps?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.