AI agent skill set for databases We're introducing database-tools - a new AI agent skill set for working with databases. Skills define reusable procedures for AI agents. They provide step-by-step instructions for completing complex tasks using existing DataGrip features, including MCP tools. This helps agents generate higher-quality results at a lower cost. The skill set is bundled with DataGrip. To ensure the skill set is enabled, go to Settings | Tools | AI Assistant | Skills. Data source management with MCP tools We're expanding the list of MCP tools that assist AI agents in working with databases more efficiently and help reduce costs. In this release, we are expanding the list of tools with two new ones: create_database_connection and edit_database_connection. These will help agents with the following tasks:

• Creating new data sources
• Running test connections
• Diagnosing connection issues
• Updating existing data source settings To improve how agents use these tools, we have described them in the database-tools skill set. To have an agent create a new data source, ask it to do so and provide the connection details. The agent will create the data source and run a test connection. The data source settings contain the connection details you provided. If you need the agent to diagnose a connection issue or update an existing data source, ask it to do so, providing the details if needed. Improvements to MCP tools and the AI Chat tool window
• Previously, AI agents received a generic "not done" error when an SQL query failed. Now, the agent receives an actual database message with an SQL error.
• The list_database_connections MCP tool now returns database connection information correctly. Previously, incorrect handling of parameters with default values caused an output schema error.
• The database schema attachment UI in the AI Chat tool window has been redesigned. User consent for database operations in the CLI When you work with an agent in a command-line interface (CLI) and the agent needs to perform an operation on a database, it now asks for user consent and shows what it is about to do. The consent form supports syntax highlighting for SQL queries. Note that for multiline queries in Claude Code, the options at the bottom of the form might not be displayed properly. The current workaround is to resize the terminal window or switch the terminal UI to the default mode by running /tui default. The Explain Plan raw format actions are back In the previous release, we simplified the Explain Plan dropdown by removing the Explain Plan (Raw) and Explain Analyse (Raw) actions. This change affected workflows that regularly require these actions. In this release, we have addressed your feedback by bringing back the actions and making them accessible from the toolbar.
• We have added a Raw button to the left-hand toolbar of the Query Plan tab. For the explain plan tree, the button opens the raw explain plan in the Raw inner tab. The explain plan tree also remains available in the Operation Tree inner tab.
• We have brought back the Explain Plan (Raw) and Explain Analyse (Raw) actions. To have them in the Explain Plan dropdown, go to Settings | Appearance & Behavior | Menus and Toolbars, select Explain Plan, and click Add. Then, select the actions from the list and click OK. Place the actions in your preferred order. They will then appear in the Explain Plan dropdown.
• When you open an explain plan in raw format by using Explain Plan (Raw), the plan opens in the Query Plan tab, too. For this format, the toolbar on the left-hand side of the tab has the Show Tree button instead of Raw. This button opens the Operations Tree inner tab with the explain plan tree. The raw plan remains available in the Raw tab. If you use Explain Plan and Explain Plan (Raw) actions on the same query, DataGrip opens the corresponding inner tabs next to each other in the Query Plan tab. Bundled JDBC drivers Previously, DataGrip didn't include any JDBC drivers. Each driver you wanted to use had to be downloaded separately when creating a data source. Now, the following JDBC drivers are bundled in DataGrip:
• Microsoft SQL Server ver. 13.2.1
• MongoDB ver. 1.21
• MySQL Connector/J ver. 9.5.0
• PostgreSQL ver. 42.7.3
• Redis ver. 1.6 To learn more about using JDBC drivers in DataGrip, refer to this documentation topic. Deletion of folders and their contents Previously, deleting non-empty folders was not supported. Now, you can delete folders that contain other folders or data sources. Before deleting a folder, DataGrip will ask for confirmation and display the list of items to be deleted. Improved support for ClickHouse and Snowflake SQL We've expanded ClickHouse SQL support with the following additions:
• Query parameter syntax.
• Additional SYSTEM statements.
• The FIRST clause in ALTER TABLE ADD COLUMN statements. Other enhancements include:
• Improved support for ORDER BY... WITH FILL STEP.
• Fixes for false syntax errors related to projection indexes. For Snowflake, we've added support for:
• Flow operations.
• DATABASE ROLE syntax.
• HASH functions, including HASH(*).
• INTERSECT, UNION BY NAME, and DISTINCT in SELECT statements.
• Object, session, and user parameters in ALTER USER statements.