octocode
MCP server for semantic code research and context generation on real-time using LLM patterns | Search naturally across public & private repos based on your permissions | Transform any accessible codebase/s into AI-optimized knowledge on simple and complex flows | Find real implementations and live docs from anywhere
claude mcp add --transport stdio bgauryy-octocode-mcp npx octocode-mcp@latest
How to use
Octocode MCP Server provides an AI-enabled coding assistant that connects your AI agent to code repositories and local tooling. It exposes capabilities for deep code exploration via LSP, GitHub and GitLab integration (search repositories, identify usage patterns, read implementations, explore PRs), and local workspace navigation (search code, browse directories, locate files) with semantic navigation. The server is designed to empower an AI assistant with research-driven skills, enabling workflows like evidence-based exploration, plan-driven execution, and PR review across multiple domains. Once running, you can invoke its skills to perform tasks such as researching code bases, performing local searches, planning implementations, and reviewing PRs with structured, evidence-backed output. The server’s design emphasizes integration with existing tooling (GitHub/GitLab, LSP, local file systems) and a suite of skills to extend agent capabilities, including Research, Local Search, Plan, PR Reviewer, Roast, Prompt Optimizer, and Documentation Writer.
How to install
Prerequisites:
- Node.js (preferred 18.x or newer) and npm installed on your system
- GitHub authentication (for seamless access to private repos and the authentication setup described in the docs)
Installation steps:
- Verify Node.js and npm are installed
node -v
npm -v
- Run the Octocode MCP server via npx (recommended via Octocode CLI)
npx octocode-cli
- Alternative: manually configure the MCP server in your MCP config
{
"mcpServers": {
"octocode": {
"command": "npx",
"args": ["octocode-mcp@latest"]
}
}
}
- If you want to install a specific skill directly (example for Research):
npx add-skill https://github.com/bgauryy/octocode-mcp/tree/main/skills/octocode-research
- Start interacting with the server via your MCP host tooling as described in the documentation references.
Additional notes
- The MCP server relies on GitHub authentication for full access to repositories and advanced discovery features; ensure authentication setup is completed as per the Authentication Setup guide.
- You can install and use the server through the recommended Octocode CLI (npx octocode-cli) for an interactive setup experience that handles OAuth, MCP server installation, and a skills marketplace.
- If you’re using the manual configuration approach, keep your MCP’s global configuration in sync with the octocode entry to ensure the server boots with the correct command and arguments.
- The npm package name for this server (when using npx) is octocode-mcp; you can reference it in automation scripts as the package to install or run.
- For local testing, you can combine the Octocode MCP server with Local Tools + LSP Reference and Skills guides to understand how each capability is exposed to your AI agent.
- If you encounter issues with license, authentication, or API quotas, consult the Troubleshooting and Documentation sections of the repository.
Related MCP Servers
bytebot
Bytebot is a self-hosted AI desktop agent that automates computer tasks through natural language commands, operating within a containerized Linux desktop environment.
cursor-talk-to-figma
TalkToFigma: MCP integration between AI Agent (Cursor, Claude Code) and Figma, allowing Agentic AI to communicate with Figma for reading designs and modifying them programmatically.
mcp-shrimp-task-manager
Shrimp Task Manager is a task tool built for AI Agents, emphasizing chain-of-thought, reflection, and style consistency. It converts natural language into structured dev tasks with dependency tracking and iterative refinement, enabling agent-like developer behavior in reasoning AI systems.
OpenContext
A personal context store for AI agents and assistants—reuse your existing coding agent CLI (Codex/Claude/OpenCode) with built‑in Skills/tools and a desktop GUI to capture, search, and reuse project knowledge across agents and repos.
Context-Engine
Context-Engine MCP - Agentic Context Compression Suite
claude-code
Model Context Protocol (MCP) servers with Claude Code. These tools dramatically enhance Claude Code's capabilities, allowing it to interact with your filesystem, web browsers, and more.