JetBrains released a video walkthrough of the AI agent integration in the latest DataGrip release, supporting Claude Code, Codex, and Junie via built-in MCP tools and skills. The video covers setting up connections from text descriptions or JDBC URLs, asking agents natural-language questions about schema architecture, text-to-SQL querying, AI-driven schema cleanup with dependency safety checks, and targeting specific objects with @dbObject and @fileName mentions.
Questions this post answers
How do I use Claude Code, Codex, or Junie with DataGrip to query my database?
DataGrip provides built-in MCP tools and skills that power up AI agents like Claude Code, Codex, and Junie with database capabilities. You can set up connections from a text description or JDBC URL, ask agents about your schema in natural language, run text-to-SQL queries, and target specific objects using @dbObject or files with @fileName identifiers. Developers wiring AI agents into their database workflow can follow discussions on tools like this via daily.dev.
Can AI agents in DataGrip automatically clean up unused database tables?
Yes, DataGrip's AI agent integration includes a schema cleanup capability that auto-detects out-of-place tables and performs dependency safety checks before executing any cleanup operations, reducing the risk of accidentally breaking dependent objects during automated schema maintenance. Teams evaluating AI-assisted schema maintenance tools can track updates like this on daily.dev.