nao：搭建可评测的分析智能体

👾 nao is an open source analytics agent. (1) Create context with nao-core cli, (2) deploy nao chat interface for everyone

nao The #1 Open-Source Analytics Agent 🌐 Website · 📚 Documentation · 💬 Slack What is nao? nao is a framework to build and deploy analytics agent. Create the context of your analytics agent with nao-core cli: data, metadata, modeling, rules, etc. Deploy a UI for anyone to chat with your agent and run analytics on your data. Key Features For data teams: 🧱 Open Context Builder — Create a file-system like context for your agent. Add anything you want in the context: data, metadata, docs, tools, MCPs. No limit. 🏳️ Data Stack Agnostic — Works with any data warehouse, stack, type of context, LLM. 🕵🏻‍♀️ Agent Reliability Visibility — Unit test your agent performance before deploying it to users. Version the context and track the performance of your agent over time. Get users feedbacks to improve the agent and track their usage. 🔒 Self-hosted & secure — Self-host your analytics agent and use your own LLM keys to guarantee maximum security for your data. For business users: 🤖 Natural Language to Insights — Ask questions in plain English, get analytics straight away 📊 Native Data Visualization — Create and customize visualizations directly in the chat interface 🧊 Transparent Reasoning — See the agent reasoning and sources clearly 👍 Easy Feedback — Send feedback to the data team when a answer is right or wrong ⚡️ Quickstart your agent in 1 minute Step 1: Install nao-core package pip install nao-core Step 2: Initialize a nao project nao init It will ask you: To name your project If you want to connect a database (optional) If you want to add a repo in agent context (optional) To add an LLM key (optional) If you want to setup a Slack connection (optional) 💡 You can skip any optional question and configure them later in your nao_config.yaml file. This will create: A new folder with your project name An architecture for your context files A nao_config.yaml configuration file A RULES.md file Step 3: Verify your setup cd to the project folder and run: nao debug Step 4: Synchronize your context nao sync This will populate your context folder with your context files (data, metadata, repos, etc.) Step 5: Launch the chat and ask questions nao chat This will start the nao chat UI. It will open the chat interface in your browser at http://localhost:5005. From there, you can start asking questions to your agent. Evaluation framework Unit test your agent performance before deploying it to users. First, create a folder tests/ with questions and expected SQL in yaml. Then, measure agent's performance on examples with nao test command: nao test View results in tests panel: nao test server Commands nao --help Usage: nao COMMAND ╭─ Commands ────────────────────────────────────────────────────────────────╮ │ chat Start the nao chat UI. │ │ init Initialize a new nao project. │ │ sync Sync context from your context sources (databases, repos) │ │ test Measure agent's performance on test examples. │ │ debug Debug and troubleshoot your nao setup. │ │ --help (-h) Display this message and exit. │ │ --version Display application version. │ ╰───────────────────────────────────────────────────────────────────────────╯ 🐳 Docker Pull the image from DockerHub: docker pull getnao/nao:latest Run nao chat with Docker using the example project bundled in the image: docker run -d \ --name nao \ -p 5005:5005 \ -e BETTER_AUTH_URL=http://localhost:5005 \ getnao/nao:latest Run nao chat with Docker using your local nao project: docker run -d \ --name nao \ -p 5005:5005 \ -e BETTER_AUTH_URL=http://localhost:5005 \ -v /path/to/your/nao-project:/app/project \ -e NAO_DEFAULT_PROJECT_PATH=/app/project \ getnao/nao:latest Access the UI at http://localhost:5005 (or at any URL you configured). See the DockerHub page for more details. For end-to-end self-hosted deployment (for example on Cloud Run with PostgreSQL), see the Deployment Guide. 🧩 Skills Use nao skills to create your nao project and context with your favourite agent. Install the nao skill pack: npx skills add getnao/nao Docs on nao skills: https://docs.getnao.io/nao-agent/context-engineering/skills 👩🏻‍💻 Development See CONTRIBUTING.md for development setup, commands, and guidelines. Start nao locally Prerequisites: nvm bun uv Run the following commands: npm run npm:pin npm install npm run db:migrate -w apps/backend Create a .env file at the root of the repository and input environment variables Minimum variables necessary are: BETTER_AUTH_SECRET=#run `openssl rand -base64 32` to generate # Public URL of your app (used for authentication, trusted origins, and Slack redirects) BETTER_AUTH_URL=http://localhost:3000 # LLM providers OPENAI_API_KEY=12345678 NAO_DEFAULT_PROJECT_PATH=#path to example folder Run the following command: npm run dev Go to http://localhost:3000 on your browser 📒 Stack Backend Fastify: https://fastify.dev/docs/latest/ Drizzle: https://orm.drizzle.team/docs/get-started tRPC router: https://trpc.io/docs/server/routers Frontend tRPC client: https://trpc.io/docs/client/tanstack-react-query/usage Tanstack Query: https://tanstack.com/query/latest/docs/framework/react/overview Shadcn: https://ui.shadcn.com/docs/components ⛹️‍♀️ Join the Community Star the repo Subscribe to releases (Watch → Custom → Releases) Follow us on LinkedIn Join our Slack Contribute to the repo! 🫰🏻 Partners nao Labs is a proud Y Combinator company! 📄 License This project is licensed under the Apache 2.0 License - see the LICENSE file for details.