A local-first autonomous data analysis agent. Ask real questions about your data; it investigates — looking, computing, revising its approach when the data disagrees with it — then verifies the result and explains it, streaming its reasoning as it goes.
Your data never leaves your machine unless you choose a remote cloud provider. Powered by a modular Tri-Model Architecture (Manager, Worker, and Embeddings). No API key is required — local Ollama or LM Studio models running locally are all it takes.
📖 Documentation · 🚀 Quickstart · 🌐 Live Website · 💬 Discussions · 📦 Latest Release
brew install Wizard-AIA/wizard/wizardLinux & macOS:
curl -fsSL https://wizardw2.vercel.app/install.sh | shWindows (PowerShell 5.1 or 7):
irm https://wizardw2.vercel.app/install.ps1 | iexThen, from any directory:
wizard --version # confirm the install
wizard init # choose a provider and models, install what is missing
wizard start # launches Wizard and opens http://localhost:3000
wizard doctor # if anything looks wrong: checks the install and says how to fix it
git clone https://github.com/Wizard-AIA/Wizard-w2.git
cd Wizard-w2
docker compose up -dPrebuilt archives for macOS (Apple Silicon and Intel), Linux (x86-64 and ARM64) and Windows (x86-64), with a SHA256SUMS file, are on the latest release page. Check the archive against SHA256SUMS, extract it, then from the extracted Wizard-v<version>-<platform> folder run ./cli/wizard init and ./cli/wizard start.
Open http://localhost:3000 to begin analyzing your data. Nothing above needs administrator rights.
| Repository | Description |
|---|---|
| Wizard-w2 | Core Engine: FastAPI backend, analytical workspace, Apache Arrow streaming, OS sandboxing, and Go CLI supervisor. |
| website | Docs & Landing: Official website, architectural documentation hub, and interactive web demo. |
| homebrew-wizard | Package Manager: Official Homebrew tap for one-command macOS and Linux installation (brew install wizard). |
| skills | Skill Registry: Community registry of curated SKILL.md domain modules the agent retrieves and executes mid-turn. |
| awesome-wizard | Ecosystem Hub: Curated playbooks, sample datasets, local model recipes, and community integrations. |
| .github | Community & Health: Organization profile landing page, issue templates, and security policies. |
Thank you to everyone building and improving Wizard!
BSD-3-Clause Licensed. Created by Aniket Saha (@Aniket-a14). Contributions are welcome — see CONTRIBUTING.md to get involved.