db9 introduces a serverless PostgreSQL database designed specifically for AI agents, combining traditional SQL power with built-in file system operations. The platform eliminates the need for separate vector databases and embedding pipelines by integrating auto-embeddings, vector search, and HTTP extensions directly into SQL queries. Developers can now run similarity searches and call external APIs without leaving the database.

The tool supports branching environments where entire workspaces including data, files, cron jobs, and user permissions can be cloned with one command. This enables testing against real conditions before deletion. Built-in file storage removes the complexity of S3 configuration, allowing direct upload, download, and mounting of files alongside database tables. Distributed cron scheduling ensures jobs run reliably without idle timeouts.

With TypeScript and Python type generation built in, db9 offers zero-configuration setup that passes 600+ ORM tests. The platform works seamlessly with popular AI coding tools like Claude Code, Cursor, and VS Code, allowing agents to autonomously install, authenticate, and manage databases. By unifying structured data in Postgres with unstructured context in files, db9 provides AI agents a single workspace for everything from personal assistants to multi-agent automation runs.