LLM Context

v1.0.0Coding Agentsstable

Share code with LLMs via Model Context Protocol or clipboard. Rule-based customization enables easy switching between different tasks (like code review and documentation). Includes smart code outlining.

claude-desktopclicodingmodel-context-protocoltools
Share:
300
Stars
0
Downloads
0
Weekly
0/5

What is LLM Context?

LLM Context is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to share code with llms via model context protocol or clipboard. rule-based customization enables easy switching between different tasks (like code review and documentation). includes smart code outlinin...

Share code with LLMs via Model Context Protocol or clipboard. Rule-based customization enables easy switching between different tasks (like code review and documentation). Includes smart code outlining.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Share code with LLMs via Model Context Protocol or clipboard

Use Cases

Share code with LLMs via Model Context Protocol with rule-based customization.
Switch contexts between tasks like code review and documentation automatically.
Generate smart code outlines for better context management.
cyberchitta

Maintainer

LicenseApache-2.0
Languagepython
Versionv1.0.0
UpdatedMay 17, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx llm-context-py

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use LLM Context

LLM Context (llm-context.py) is a Python tool that packages focused, rule-filtered slices of your codebase and delivers them to AI coding assistants either via the clipboard or through an MCP server connection. It uses composable rule profiles stored in .llm-context/ to define exactly which files to include for a given task — so the context for a code review differs from the context for writing documentation. The MCP tools lc_outlines, lc_preview, and lc_missing let Claude request project structure and individual files on demand, dramatically reducing token waste compared to blindly pasting entire directories.

Prerequisites

  • Python 3.10+ and the uv package manager (https://github.com/astral-sh/uv)
  • Claude Desktop, Claude Code CLI, or another MCP-compatible client
  • A local code repository — run lc-init from the project root before first use
  • No external API keys required
1

Install llm-context with uv

Install the package as a uv tool so the lc-mcp, lc-init, lc-select, and lc-context commands are globally available.

uv tool install "llm-context>=0.6.0"
2

Initialize your project

Navigate to your repository root and run lc-init. This scaffolds the .llm-context/ configuration directory and copies skill files to .claude/skills/.

cd /path/to/your/repo
lc-init
3

Define or review rule profiles

Edit the rule files inside .llm-context/ to control which files each named profile includes. Run lc-preview with a profile name to check its coverage before using it.

lc-preview code-review
4

Add the MCP server to Claude Desktop

Register the lc-mcp command in your claude_desktop_config.json. Claude will then have access to the lc_outlines, lc_preview, and lc_missing tools for your projects.

5

Restart Claude Desktop and test the connection

Restart Claude Desktop after editing the config. Open a conversation, navigate to your project folder, and ask Claude to call lc_outlines to verify the server responds correctly.

LLM Context Examples

Client configuration

Add this to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) to register the llm-context MCP server.

{
  "mcpServers": {
    "llm-context": {
      "command": "uvx",
      "args": ["--from", "llm-context", "lc-mcp"]
    }
  }
}

Prompts to try

Use these prompts after the MCP server is connected to leverage project context tools.

- "Call lc_outlines to give me a structural overview of this project"
- "Use lc_preview with the 'documentation' rule to show me what files that profile covers"
- "Fetch the contents of src/auth/session.py using lc_missing"
- "Switch to the code-review rule profile and then review the recent changes to the API layer"

Troubleshooting LLM Context

lc-init reports permission errors or fails to create .llm-context/

Ensure you have write permissions to the project root directory and that you are running lc-init from within the repository, not from a parent directory.

lc_outlines returns an empty result in Claude

Verify the .llm-context/ directory exists in your project root and that the active rule profile's file globs match at least some files. Run 'lc-preview <rule>' in the terminal to debug the selection.

Claude Desktop cannot start the lc-mcp server

Confirm uvx is on the PATH visible to Claude Desktop's process. On macOS, add 'export PATH=$HOME/.local/bin:$PATH' to ~/.zprofile and restart the system or re-launch Claude Desktop from a shell that has the updated PATH.

Frequently Asked Questions about LLM Context

What is LLM Context?

LLM Context is a Model Context Protocol (MCP) server that share code with llms via model context protocol or clipboard. rule-based customization enables easy switching between different tasks (like code review and documentation). includes smart code outlining. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install LLM Context?

Follow the installation instructions on the LLM Context GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with LLM Context?

LLM Context works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is LLM Context free to use?

Yes, LLM Context is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "llm-context-py": { "command": "npx", "args": ["-y", "llm-context-py"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use LLM Context?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides