AI Assistant for Technical Products — Open Source and Developer-First

---

What is Toq?

Toq makes your documentation AI-powered and conversational. Developers can ask questions directly inside your docs, GitHub repos, wikis, or knowledge bases, and get instant, accurate answers without hunting through pages.

Embed Toq easily into any doc site — GitBook, ReadMe, Mintlify, Docusaurus, and more — and transform your documentation into an interactive AI assistant.

---

🚀 Features at a Glance

Feature	Why It Matters
🤖 AI Assistant for Docs	Provides instant, conversational answers from your content
🔎 Deep Memory	Understands context, has memory and can answer questions based on the context
📚 Multi-source Knowledge	Integrate GitHub, GitBook, Notion, Google Drive, Markdown
🔗 Universal Embedding	Widget works seamlessly with any static or dynamic docs site
🌐 Dedicated Query Endpoint	Offer ask.yourdomain.com for direct Q&A access
🛠 Open Source & Extensible	Customize connectors and behaviors easily

---

🎬 Demo Preview

mintlify-demo-web.mp4

Chat directly inside your docs for real-time answers.

---

🏁 Quick Start using Docker

npx @torqbit/toq

• What it does:

  • Checks for Docker and Docker Compose.
  • Writes a docker-compose.yml to your current directory (asks before overwriting).
  • Starts the stack with docker compose up -d --build.

• Services exposed by default:

  • Web: http://localhost:8080
  • MySQL: localhost:3360
  • Qdrant: http://localhost:6333

---

🚀 Quick Start using Source Code

Prerequisites

• MySQL running (local or Docker) with accessible connection credentials
• Qdrant vector database (Docker recommended):

  docker run -v ${pwd}/qdrant:/qdrant -p 6333:6333 -p 6334:6334 qdrant/qdrant

  Access the Qdrant dashboard at http://localhost:6333/dashboard
• Email provider credentials (e.g., Resend API key or SMTP settings) for signup/login and notifications
• OpenAI API key for generating AI responses

Run Locally

git clone https://github.com/torqbit/toq.git
cd toq
yarn install
yarn dev