Your personal AI-powered documentation search engine. Index, search, and chat with your docs using cutting-edge semantic search.
ContextStream is an open-source platform that brings the power of AI-driven semantic search to your documentation. Whether you're building an internal knowledge base, indexing your company's docs, or just want to search through your favorite frameworks' documentation faster, ContextStream has you covered.
Traditional documentation search relies on keyword matching, which often misses the mark. ContextStream uses AI embeddings and semantic search to understand what you're actually asking for - not just matching keywords, but understanding intent and context.
- 🧠 Semantic Search: Ask questions naturally, get relevant answers
- 🔌 Multiple Sources: GitHub repos, websites, wikis, and more
- 💬 AI Chat Interface: Ask questions, get answers from your docs
- 🔄 Auto-Sync: Keep your docs fresh with automatic updates
- 🎯 MCP Integration: Works with Claude and other AI coding assistants
- 🔐 Privacy-First: Self-hosted, your data stays yours
- 🚀 Fast & Efficient: Hybrid BM25 + vector search for optimal results
ContextStream is free and open-source, but hosting, development, and maintenance require time and resources. If you find this project helpful, please consider supporting it!
- Star this repo - It helps with visibility!
- Sponsor on GitHub - Become a sponsor to help keep this project alive
- Share - Tell others about ContextStream
- Contribute - Code, docs, bug reports, and ideas are all welcome!
Your logo here! Be the first to support this project and get your name/logo featured here.
Interested in sponsoring? Get in touch
Get up and running in under 5 minutes!
- Docker and Docker Compose
- OpenAI API key (get one here)
- 2GB RAM minimum (4GB recommended)
# Download docker-compose file
curl -o docker-compose.yml https://raw.githubusercontent.com/jimseiwert/context-stream/main/docker/docker-compose.production.yml
# Create environment file
cat > .env << EOF
OPENAI_API_KEY=your-api-key-here
BETTER_AUTH_SECRET=$(openssl rand -base64 32)
NEXTAUTH_SECRET=$(openssl rand -base64 32)
EOF
# Start ContextStream
docker-compose up -d
# Visit http://localhost:3000 🎉That's it! You now have your own AI-powered documentation search engine running locally.
- Developer Teams: Index your internal wikis, API docs, and codebases
- Content Teams: Search through marketing materials, guides, and knowledge bases
- Students/Researchers: Build a personal search engine for your research papers and notes
- Open Source Maintainers: Make your project's docs more discoverable
- AI Coding Assistants: Use with Claude Code or other AI tools via MCP
Choose the deployment that fits your needs:
Production-ready setup with separate services for better performance.
Enterprise-ready with auto-scaling and high availability.
Built with modern, battle-tested technologies:
- Frontend: Next.js 15, React 19, TailwindCSS, shadcn/ui
- Backend: Next.js API Routes, Drizzle ORM
- Database: PostgreSQL 16 with pgvector
- Search: Hybrid BM25 + Vector similarity
- Queue: Bull (Redis-based)
- Auth: Better Auth with OAuth support
- AI: OpenAI Embeddings & GPT-4
- Deployment: Docker, Kubernetes
Uses hybrid search combining traditional BM25 with AI vector embeddings for the best of both worlds.
Organize your docs into workspaces - personal, team, or global. Control access and permissions.
Built-in Model Context Protocol server for seamless integration with Claude Code and other AI coding assistants. Search your docs without leaving your editor!
Keep your docs fresh with automatic synchronization. Set schedules or use webhooks for instant updates.
Clean, responsive interface built with Next.js and Tailwind. Works great on desktop and mobile.
We ❤️ contributions! Whether you're fixing bugs, adding features, improving docs, or just discussing ideas, you're welcome here.
- Fork the repository
- Create a feature branch:
git checkout -b feature/amazing-feature - Commit your changes:
git commit -m 'Add amazing feature' - Push to the branch:
git push origin feature/amazing-feature - Open a Pull Request
Check out our Contributing Guide for more details.
New to the project? Look for issues tagged with good first issue - these are great starting points!
Want to contribute or run locally? Here's how:
# Clone the repo
git clone https://github.com/jimseiwert/context-stream.git
cd context-stream
# Install dependencies
npm install
# Start dependencies (PostgreSQL, Redis)
docker-compose up -d postgres redis
# Setup environment
cp .env.example .env
# Edit .env with your values
# Run migrations
npm run db:migrate
# Start dev server
npm run dev
# In another terminal, start the worker
npm run workerVisit http://localhost:3000 and start hacking! 🎉
- 🚀 Deployment Guide - Complete deployment instructions
- ⚙️ Configuration - Environment variables and settings
- 🐳 Docker Images - Available images
- ☸️ Helm Chart - Kubernetes deployment
- 🔌 API Docs - REST API reference
- 🤖 MCP Integration - Model Context Protocol setup
Exciting features on the horizon:
- Multi-modal search (images, diagrams, code screenshots)
- Local embedding models (no OpenAI dependency)
- More data sources (Notion, Confluence, Google Docs)
- Advanced analytics and insights
- Browser extension for quick searches
- Mobile app
- Plugin system for custom integrations
- Self-hosted LLM support (Ollama, LM Studio)
Have an idea? Open an issue or start a discussion!
- Stars: ⭐ Star us on GitHub
- Docker Pulls: 🐳 Check out our Docker Hub
- Contributors: 👥 See all our amazing contributors
Join our community and get help:
- Discussions: GitHub Discussions - Ask questions, share ideas
- Issues: GitHub Issues - Bug reports and feature requests
- Email: support@contextstream.dev - Direct support
- Discord: Coming soon! 🎮
ContextStream wouldn't be possible without these amazing open-source projects:
- Next.js - The React framework
- Drizzle ORM - Lightweight TypeScript ORM
- pgvector - Vector similarity search for PostgreSQL
- Bull - Premium queue package
- Better Auth - Modern authentication
- TailwindCSS - Utility-first CSS framework
- shadcn/ui - Beautiful UI components
And a huge thanks to all our contributors! 🎉
ContextStream is open-source software licensed under the MIT License.
You're free to use, modify, and distribute this software. See the LICENSE file for full details.
Made with ❤️ by the open-source community