AI-powered API explorer for Databricks — browse, test, and analyze 500+ REST APIs in one interface.
Intelligence Studio is an open-source platform for exploring, testing, and integrating with Databricks REST APIs. It combines a visual API explorer with AI-powered assistance — natural language search, response analysis, code generation, SQL execution, and interactive data visualization.
No more switching between documentation, curl commands, and Postman.
- 640+ API Endpoints — Complete Databricks REST API catalog organized by service (Compute, Unity Catalog, Jobs, SQL, ML, Account APIs, and more)
- 13 AI Assistant Features — Data QA, endpoint discovery, response analysis, error diagnosis, code generation, workflow builder, agent chat, and more
- SQL Query Editor — Built-in editor with Unity Catalog browser, syntax highlighting, and auto-visualization
- Data Visualization — 12 chart types with auto-detection, dashboards, schema visualizer, and dependency graphs
- Code Generation — Export working code in Python, cURL, JavaScript, TypeScript, Go, and PowerShell
- Azure Multi-Workspace Login — OAuth-based authentication with multi-tenant, multi-subscription support
- Full Catalog Export — Export all 640+ endpoints as Postman Collection, OpenAPI 3.0, or Insomnia with folder structure and request bodies
- Export Everywhere — Postman, Insomnia, OpenAPI, GitHub Actions, PDF, Word, Excel, CSV, JSON
- Request History — Full replay capability with favorites
- Cross-Platform — Web app, macOS desktop, Windows desktop, and Python CLI
| API Explorer | AI Analysis |
|---|---|
 |
 |
| Visualizations | Agent Chat |
|---|---|
 |
 |
| Settings | Test Data |
|---|---|
 |
 |
Pre-built binaries are available on the Releases page.
| Platform | File | How to run |
|---|---|---|
| macOS (Apple Silicon) | Intelligence-Studio-1.0.0-arm64-mac.zip |
Unzip and move to Applications |
| Windows (Installer) | Intelligence-Studio-Setup-1.0.0.exe |
Run the installer |
| Windows (Portable) | Intelligence-Studio-1.0.0.exe |
Double-click to run — no install needed |
| Web | — | Clone the repo and run make dev |
| CLI | — | Clone the repo and run make cli-install |
macOS only: Right-click the
.app→ Open to bypass Gatekeeper on first launch (or runxattr -d com.apple.quarantine /path/to/"Intelligence Studio.app"in Terminal).
Open the app and go to Settings to complete the following before use:
- Enter your Databricks workspace URL (e.g.
https://<workspace>.azuredatabricks.net) - Enter your Personal Access Token (PAT)
- Click Connect
- In Settings, select your SQL Warehouse from the dropdown
- Required for running queries and data exploration
- In Settings, select a Databricks Foundation Model endpoint
- Powers the AI Assistant, Data QA, and analysis features
- Requires a model serving endpoint enabled in your Databricks workspace
- Node.js 18+ and npm 9+
- Python 3.11+
# Clone the repository
git clone https://github.com/viral0216/Intelligence-Studio.git
cd Intelligence-Studio
# Install dependencies
make install
# Start development servers
make devThis starts:
- Frontend on http://localhost:5173
- Backend on http://localhost:8000
Open http://localhost:5173 and follow the Post Installation steps below.
After the app is running, open Settings to complete the following before use:
- Enter your Databricks workspace URL (e.g.
https://<workspace>.azuredatabricks.net) - Enter your Personal Access Token (PAT)
- Click Connect
- In Settings, select your SQL Warehouse from the dropdown
- Required for running queries and data exploration
- In Settings, select a Databricks Foundation Model endpoint
- Powers the AI Assistant, Data QA, and analysis features
- Requires a model serving endpoint enabled in your Databricks workspace
make build-mac # macOS (.app + .zip)
make build-win # Windows (.exe installer + portable)make cli-installFrontend React 18 + TypeScript + Vite + Zustand + Tailwind CSS 4
Backend Python + FastAPI + Uvicorn + httpx + Pydantic v2
Desktop Electron 28
CLI Python Click + Rich
AI Databricks Foundation Models (Llama, Claude, Gemma)
Intelligence-Studio/
frontend/ React + TypeScript frontend
backend/ Python FastAPI backend
desktop/ Electron desktop app
cli/ Python CLI tool
scripts/ Build and utility scripts
docs/ Documentation and demo assets
For detailed documentation — full feature walkthrough, architecture, environment variables, export guides, and more — see docs/README.md.
| Metric | Value |
|---|---|
| API endpoints cataloged | 640+ |
| AI assistant features | 13 |
| Code generation languages | 6 |
| Export formats | 10+ |
| Chart types | 12 |
| Feature flags | 14 |
| Customizable AI prompts | 14 |
- Databricks tokens are never stored server-side — every request carries its own credentials
- AI-generated scripts are sandboxed with blocklist validation before execution
- All request payloads are validated with Pydantic v2 schemas
See SECURITY.md for details on reporting vulnerabilities.
Contributions are welcome! Please read CONTRIBUTING.md for setup instructions, coding guidelines, and the PR process.
This project follows the Contributor Covenant Code of Conduct.
- Adding new API endpoint presets to the catalog
- Improving AI prompt templates
- Adding new chart types and visualizations
- Writing tests (frontend and backend)
- Documentation improvements
- Accessibility enhancements
- Performance optimizations
- Multi-cloud authentication (AWS, GCP)
Thanks to everyone who contributes to Intelligence Studio!
Note: This is the first release of Intelligence Studio. You may encounter bugs or rough edges with some AI features. If you find an issue, please open it on GitHub — I maintain this project on weekends, so responses may take a few days. Contributions and feedback are always welcome!