octocode-mcp

A Model Context Protocol (MCP) server enabling AI assistants to search, analyze, and extract insights from millions of GitHub repositories with enterprise-grade security and token efficiency.

Table of Contents

• See It In Action
• Installation
• More Examples
• Overview
• Tools
• Commands
  • /research - Expert Code & Product Research
  • /plan - Research, Plan & Implement Complex Tasks
  • /review_pull_request - Comprehensive PR Review
  • /review_security - Security Audit
• Documentation
• Project Stats
• Community
• License

See It In Action

Full-Stack Application Built in Under 10 Minutes

Watch AI assistant use Octocode to research, plan, and build a complete chat application with Express backend.

Prompt:

Use Octocode MCP for Deep Research

I want to build an application with chat (front-end) that shows a chat window to the user. The user enters a prompt in the chat, and the application sends the prompt to an Express backend that uses AI to process the request.

Add a return box (to show the message returned from the AI) and loaders to the UI. I want to build an AI agent system in Node.js using LangChain and LangGraph. Can you research the latest patterns?

Please conduct thorough research on how to create this in the best way possible. Focus on repositories with good documentation and recent activity.

• Do a deep research
• Create a plan document
• Initiate the plan and create the application

Phase 1: Research & Planning

https://github.com/user-attachments/assets/4225ab98-ae2f-46dc-b3ce-7d117e552b8c

Octocode Plan Document - Detailed architecture and step-by-step guide

Phase 2: Implementation

https://github.com/user-attachments/assets/2aaee9f1-3592-438a-a633-255b5cbbb8e1

Result: Production-ready full-stack application with authentication, real-time features, and best practices - All in less than 10 minutes

Research and Build Fullstack Agentic Application with /research command in Under 10 Minutes

Why use the /research command? Instead of manually searching through repositories and piecing together information, let the AI conduct comprehensive research for you:

• 🎯 Intelligent Tool Orchestration: Automatically selects and combines the right Octocode tools (repository search, code search, file content, PR analysis, repo structure) based on your research needs
• 🧠 Smart Decision Making: Makes strategic choices throughout the research flow—when to search broadly vs. specifically, which repositories to explore, and how to validate findings
• 👥 Multi-Purpose Research: Perfect for feature discovery (product managers), code understanding (developers), bug investigation, flow analysis, planning from scratch, dependency tracking, security audits, and more
• 🔬 Specialized Workflows: Handles Technical Research (code flows), Product Research (docs+code validation), Pattern Analysis (cross-repo comparison), Bug Investigation, Architecture Mapping, API Research, Security/Auth flows, and more
• 🔍 Transparent Reasoning: Shows you exactly which tools it's using, what it's searching for, and why at each step
• 🎨 Adaptive Strategy: Works across public repos, private organizations, and specific repositories with configurable depth (overview, deep dive, or cross-repo comparison)
• 📊 Cross-Validated Results: Leverages multiple Octocode tools to verify information from different sources and perspectives
• 🚀 Actionable Insights: Delivers implementation-ready plans with code examples, not just raw information

Prompt:

/octocode/research How can I use LangChain, LangGraph, and similar open-source AI tools to create agentic flows between agents for goal-oriented tasks? Can you suggest UI frameworks I can use to build a full-stack AI application?

https://github.com/user-attachments/assets/82ed97ae-57a9-46ae-9acd-828a509e711b

Discover APIs, Frameworks, and Dive Into Internal Implementation Details

Octocode excels at both broad discovery and deep code analysis. Whether you're exploring new APIs, finding frameworks, or understanding how popular libraries work under the hood, Octocode provides comprehensive answers in seconds.

First Prompt - Broad Discovery:

list top repositories for:

• Stock market APIs (Typescript)
• Cursor rules examples
• UI for AI
• Mobile development using React
• State management for React

What happens: Octocode searches across GitHub to find the most popular and well-maintained repositories for each category, analyzing stars, activity, documentation quality, and recent updates. You get curated lists with context about each repository's strengths.

Second Prompt - Deep Implementation Analysis:

How React implemented useState under the hood?

What happens: Octocode dives into React's source code, traces the implementation flow, analyzes the relevant files (ReactHooks.js, ReactFiberHooks.js), and explains the internal mechanics including fiber architecture, hook state management, and dispatcher patterns—all with code references and detailed explanations.

The Power: Move seamlessly from discovering what exists to understanding how it works in a single conversation. No manual repository hunting or code spelunking required.

https://github.com/user-attachments/assets/c184d5d4-c9b6-40a1-a55a-41cb9b3ecc4f

Installation

Prerequisites

• Node.js >= 18.12.0
• GitHub Authentication (choose one):
  • GitHub CLI (recommended): Install from cli.github.com and run gh auth login
  • Personal Access Token: Create at github.com/settings/tokens with scopes: repo, read:user, read:org

Getting started

First, install the Octocode MCP server with your client.

Standard config works in most of the tools:

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": [
        "octocode-mcp@latest"
      ]
    }
  }
}

Note: This configuration uses GitHub CLI authentication. For Personal Access Token, see the Authentication Guide below.

"amp.mcpServers": {
  "octocode": {
    "command": "npx",
    "args": [
      "octocode-mcp@latest"
    ]
  }
}

amp mcp add octocode -- npx octocode-mcp@latest

claude mcp add octocode npx octocode-mcp@latest

codex mcp add octocode npx "octocode-mcp@latest"

[mcp_servers.octocode]
command = "npx"
args = ["octocode-mcp@latest"]

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": ["octocode-mcp@latest"]
    }
  }
}

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": [
        "octocode-mcp@latest"
      ]
    }
  }
}

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": [
        "octocode-mcp@latest"
      ]
    }
  }
}

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "octocode": {
      "type": "local",
      "command": [
        "npx",
        "octocode-mcp@latest"
      ],
      "enabled": true
    }
  }
}

# For VS Code
code --add-mcp '{"name":"octocode","command":"npx","args":["octocode-mcp@latest"]}'

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": [
        "octocode-mcp@latest"
      ]
    }
  }
}

Authentication Methods

Octocode MCP supports two authentication methods:

Option 1: GitHub CLI (Recommended)

Advantages: Automatic token management, works with 2FA, supports SSO

# Install GitHub CLI
# macOS
brew install gh

# Windows
winget install --id GitHub.cli

# Linux
# See https://github.com/cli/cli/blob/trunk/docs/install_linux.md

# Authenticate
gh auth login

Then use the standard configuration (no GITHUB_TOKEN needed).

Option 2: Personal Access Token

When to use: CI/CD environments, automation, or if GitHub CLI isn't available

1. Create a token at github.com/settings/tokens
2. Select scopes: repo, read:user, read:org
3. Add to your MCP configuration:

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": ["octocode-mcp@latest"],
      "env": {
        "GITHUB_TOKEN": "ghp_your_token_here"
      }
    }
  }
}

Security Tip: Never commit tokens to version control. Use environment variables or secure secret management.

Verify Installation

After installation, verify Octocode MCP is working:

1. Restart your MCP client completely
2. Check connection status:
   • Cursor: Look for green dot in Settings → Tools & Integrations → MCP Tools
   • Claude Desktop: Check for "octocode" in available tools
   • VS Code: Verify in GitHub Copilot settings
3. Test with a simple query:

   Search GitHub for React hooks implementations
   

If you see Octocode tools being used, you're all set! 🎉

GitHub Enterprise Support

Octocode MCP supports GitHub Enterprise Server instances with custom API URLs.

Configuration

Add the GITHUB_API_URL environment variable to your MCP configuration:

{
  "mcpServers": {
    "octocode": {
      "command": "npx",
      "args": ["octocode-mcp@latest"],
      "env": {
        "GITHUB_TOKEN": "your_token",
        "GITHUB_API_URL": "https://github.company.com/api/v3"
      }
    }
  }
}

Default: If not specified, defaults to https://api.github.com (public GitHub).

Note: Ensure your GitHub Enterprise token has the same scopes as documented in the Authentication Guide.

More Examples

Additional Demonstrations

ThreeJS Implementation Quality Comparison

Interactive Demo

Side-by-side comparison showing:

• Generic AI: Basic implementation with common patterns
• Octocode-Enhanced AI: Production-grade implementation with advanced techniques from real projects

Key Differences:

• Performance optimizations from high-performance projects
• Proper resource management patterns
• Industry-standard error handling
• Real-world edge case handling

Deep Technical Research

YouTube: React Hooks Internals

Demonstrates progressive research workflow:

1. Repository discovery (React source)
2. Structure exploration (hooks implementation)
3. Code analysis (internal mechanisms)
4. Comprehensive explanation with code references

Overview

Octocode is an agentic code research platform that bridges the gap between AI assistants and real-world code implementations. By providing structured access to GitHub's vast repository ecosystem, it enables AI systems to learn from production codebases rather than relying solely on training data.

Core Capabilities

Tools

Octocode provides five specialized research tools designed to work together for comprehensive code analysis:

🔍 githubSearchCode

Find code implementations across repositories

Search for specific code patterns, functions, or implementations across millions of repositories.

Key Features:

• Content Search: Find code inside files by keywords (AND logic)
• Path Search: Discover files/directories by name (25x faster)
• Smart Filtering: Scope by repository, path, file extension, or popularity
• Context-Rich Results: Returns code snippets with surrounding context

Common Use Cases:

• Find implementation examples: "How do popular repos implement OAuth?"
• Discover patterns: "Search for React custom hooks in vercel repos"
• Locate functions: "Find error handling patterns in Express apps"

📚 githubSearchRepositories

Discover repositories by topics and keywords

Your starting point for repository discovery - find the right projects to analyze.

Key Features:

• Topic-Based Discovery: Search by exact GitHub topics (most precise)
• Keyword Search: Find repos by name, description, or README content
• Quality Filters: Filter by stars, language, size, activity
• Sorting Options: By popularity, recency, or relevance

Common Use Cases:

• Find popular implementations: "Discover TypeScript CLI tools with >1000 stars"
• Research ecosystems: "Find all React state management libraries"
• Organization research: "List all repos from microsoft with topic 'ai'"

🗂️ githubViewRepoStructure

Explore repository directory structure

Understand how a project is organized before diving into specific files.

Key Features:

• Directory Tree: Visual representation of folder structure
• File Sizes: See file sizes to identify important components
• Depth Control: Explore 1 level (overview) or 2 levels (detailed)
• Path Targeting: Navigate directly to specific directories

Common Use Cases:

• Project overview: "Show me the structure of facebook/react"
• Find entry points: "Explore src/ directory in a monorepo"
• Understand architecture: "Navigate to the API implementation folder"

📄 githubGetFileContent

Read file contents with smart extraction

Retrieve specific content from files efficiently - full files or targeted sections.

Key Features:

• Pattern Matching: Extract sections matching specific patterns with context
• Line Range Reading: Read specific line ranges for efficiency
• Full Content Access: Get entire file when needed
• Content Minification: Automatic optimization for token efficiency

Common Use Cases:

• Read specific functions: "Get the validateUser function from auth.ts"
• Extract sections: "Show me all the middleware definitions in app.js"
• Read configuration: "Get the full package.json file"
• Analyze specific code: "Read lines 100-150 from the API handler"

🔀 githubSearchPullRequests

Analyze pull requests, changes, and discussions

Understand how code evolved, why decisions were made, and learn from production changes.

Key Features:

• PR Discovery: Search by state, author, labels, dates
• Direct Access: Fetch specific PR by number (10x faster)
• Code Diffs: Include full diff content to see what changed
• Discussions: Access comment threads and review discussions
• Merged Code: Filter for production-ready, merged changes

Common Use Cases:

• Learn from changes: "Show recent merged PRs about authentication"
• Understand decisions: "Find PRs discussing the API redesign with comments"
• Track implementations: "See how feature X was implemented with diffs"
• Expert contributions: "Find PRs by @author in the last 6 months"

Full Documentation →

Commands

Octocode MCP provides intelligent prompt commands that enhance your research workflow:

/research - Expert Code & Product Research

Powerful research prompt leveraging Octocode's full capabilities for deep code discovery, documentation analysis, pattern identification, and bug investigation. Orchestrates parallel bulk queries with staged analysis to uncover insights fast.

When to use:

• Understanding repository workflows: Discover how repositories work, trace specific flows through codebases, and understand technical implementations
• Cross-repository flow analysis: Understand complex flows that span multiple repositories, trace data flows across microservices
• Deep technical investigations: Trace code flows, understand complex implementations, analyze architecture decisions
• Bug investigation: Find root causes by analyzing code, commit history, and related PRs
• Pattern discovery: Compare implementations across multiple repos to find best practices
• Documentation validation: Verify docs match actual code behavior

Usage Examples:

/research How does React's useState hook work internally?
/research Compare state management approaches: Redux vs Zustand vs Jotai
/research Why is the payment webhook failing? Trace the error through payment-service

/plan - Research, Plan & Implement Complex Tasks

Your AI architect for tackling complex development work. Breaks down ambitious tasks into actionable steps, researches existing patterns and implementations, then guides you through execution—all powered by Octocode's deep codebase intelligence.

When to use:

• Building new features: Research patterns, plan architecture, then implement
• Complex refactoring: Understand current state, plan migration path, execute safely
• Learning new technologies: Research best practices, create learning plan, build incrementally
• System design: Explore existing implementations, design your approach, validate decisions

Usage Examples:

/plan Build a real-time chat application with WebSocket support
/plan Migrate our authentication from JWT to OAuth2
/plan Implement a plugin system for our CLI tool

/review_pull_request - Comprehensive PR Review

Args: prUrl (required) - GitHub Pull Request URL (e.g., https://github.com/owner/repo/pull/123)

Expert-level PR review with a Defects-First mindset. Dives deep into code changes, spots bugs before they ship, flags complexity risks, and delivers actionable feedback that elevates code quality.

What it analyzes:

• Defects & Bugs: Logic errors, edge cases, race conditions, null handling
• Security Issues: Injection vulnerabilities, auth bypasses, data exposure
• Performance: N+1 queries, memory leaks, inefficient algorithms
• Code Quality: Complexity, maintainability, test coverage gaps
• Best Practices: Design patterns, error handling, documentation

Usage:

/review_pull_request prUrl: https://github.com/facebook/react/pull/12345

/review_security - Security Audit

Args: repoUrl (required) - GitHub repository URL (e.g., https://github.com/owner/repo)

Comprehensive security analysis of a repository. Identifies vulnerabilities, reviews authentication/authorization patterns, checks for secrets exposure, and provides remediation guidance.

What it analyzes:

• Authentication & Authorization: Auth flows, session management, access controls
• Input Validation: Injection points, sanitization, boundary checks
• Secrets Management: Hardcoded credentials, API keys, configuration security
• Dependencies: Known vulnerabilities, outdated packages, supply chain risks
• Data Protection: Encryption, PII handling, data flow security

Usage:

/review_security repoUrl: https://github.com/your-org/your-repo

Tips for Using Commands

1. Use /research for code exploration - Deep dive into how things work
2. Use /plan for building - Research, plan, then implement complex features
3. Use /review_pull_request before merging PRs for thorough code review
4. Use /review_security for security audits of repositories

💡 Pro Tip: Combine /research and /plan for maximum effectiveness—research existing patterns first, then plan your implementation with confidence.

Documentation

Comprehensive Guides

Local Research

octocode-mcp-local brings the power of Octocode's research capabilities to your local filesystem.

An MCP server that provides AI assistants with native Unix tools for local code exploration. Built on ripgrep for pattern search, find for file metadata queries, and ls for directory traversal.

• Fast Pattern Search: Uses ripgrep for high-performance regex searching
• Smart Navigation: Directory structure analysis and file discovery
• Secure Access: Strictly scoped to your workspace with automatic secrets filtering
• No Indexing: Direct filesystem access for real-time results

Quick Install

{
  "mcpServers": {
    "octocode-local": {
      "command": "npx",
      "args": [
        "octocode-mcp-local@latest"
      ]
    }
  }
}

👉 Get Started with Octocode Local

Community

Get Support

• GitHub Discussions: Ask questions, share ideas
• GitHub Issues: Report bugs, request features

Show Your Support

If Octocode helps your AI development workflow:

• Star the repository on GitHub
• Share on social media with #OctocodeMCP

License

MIT - See LICENSE for details.