Bring Databricks into Kiro IDE with the AI Dev Kit Power

Bring Databricks into Kiro IDE with the AI Dev Kit Power | Databricks Blog Skip to main content

Summary

Two paths to connect Kiro IDE to Databricks: the four Databricks-managed MCP servers (Genie, SQL, Unity Catalog Functions, Vector Search) for a 10-minute PAT-based setup, or the new Databricks AI Dev Kit Power — one click, all essential tools and skills, four auth options.

AI-assisted development grounded in real workspace metadata: both paths inherit Unity Catalog row, column, and tag-based grants, so the assistant writes SQL with your actual columns and only sees what you can see — no hallucinations, no unauthorized reads.

Pick by surface area: Path A is the lightest setup for analysts and SQL-first builders; Path B opens the full Databricks platform (pipelines, jobs, Mosaic AI, Agent Bricks, Lakebase, Asset Bundles) inside the IDE.

Why this matters AI-assisted development falls apart the moment the assistant has to guess at column names, table layouts, or which catalogs you can read. The fix is grounding: connect the assistant to live workspace metadata via Model Context Protocol (MCP), and the SQL it writes uses the actual columns you have, dbt models join real tables, and every query inherits the Unity Catalog grants you already have in place. Nothing leaves the platform. The AI sees only what you can see. Two milestones just landed that make this practical in Kiro IDE: First, the Databricks AI Dev Kit added Kiro support upstream in PR #511. The unified installer treats kiro as a first-class target alongside claude , cursor , copilot , codex , and gemini . One command, and Kiro picks up the full toolkit at ~/.kiro/skills/ and ~/.kiro/settings/mcp.json . Second, the Databricks AI Dev Kit Power shipped in the Kiro Powers catalog in PR #129. Open the Powers panel, click Try, and the Power runs the entire onboarding: installer, MCP wiring, auth detection, and skill loading. Combined with the four Databricks-managed remote MCP servers that already ship inside the platform, you have two ways to wire Kiro into Databricks. Both share a common outcome: builders ship analytics, pipelines, and agent workflows faster when the assistant inherits real workspace permissions instead of guessing at schemas, columns, and grants. Why Databricks for AI-assisted development The two milestones above make Kiro × Databricks practical. The reason it matters is what's underneath. Three things make Databricks the substrate of choice for AI-assisted development, regardless of which path you take. Unity Catalog is the only governance layer that grounds AI at the data level. Every MCP call — Path A or Path B — inherits row, column, and tag-based grants. The assistant doesn't have a privileged view of your data; it sees exactly what you can see. There is no separate access-control layer to manage, and no risk of the AI writing queries against tables it shouldn't even know exist. One copy of data, one set of definitions. Because Databricks is a lakehouse, the table the assistant queries through databricks-sql is the same table your dbt model writes to, the same table your Genie space exposes, the same table your AI/BI dashboard reads from. There is no warehouse-to-lake sync to break, no separate semantic layer to keep in sync. When the assistant grounds itself in samples.tpch.lineitem, it's grounding in the same definition every other tool uses. The full AI stack is integrated, not bolted on. Mosaic AI Gateway routes model calls. Agent Bricks orchestrates multi-agent workflows. MLflow tracks experiments and evaluations. Vector Search powers semantic retrieval. Lakebase handles transactional state. All of these surface in the Power, all on the same UC. You're not stitching together five products; you're using one platform. There's a fourth thing worth naming: the Power itself is Databricks-built. No other data platform ships a one-click IDE Power for Kiro, Cursor, Claude, Copilot, Codex, and Gemini. The MCP layer is open, the protocol is open, the integration is open — but the experience that wraps it is engineered by Databricks specifically for the way our customers build. The two paths at a glance Dimension Path A: Managed MCP Servers Path B: Databricks AI Dev Kit Power

Surface area 4 servers: Genie, SQL, UC Functions, Vector Search All essential Databricks tools and skills

What you get Natural-language SQL, semantic search, governed function execution Path A surface plus pipelines, jobs, dashboards, Lakebase, Mosaic AI, Agent Bricks, Asset Bundles, MLflow, model serving, Apps

Hosting Databricks-managed (remote HTTPS) Local Python MCP server via the AI Dev Kit installer

Auth PAT in shell env OAuth U2M (recommended), OAuth M2M, .databrickscfg profile, or PAT

Setup Edit  ~/.kiro/settings/mcp.json , export env vars One-click Power install plus guided auth flow

Best for Analysts and SQL-first builders who want a 10-minute path to ask their warehouse a question Data engineers and platform builders who need the full Databricks surface area in one IDE

Integration architecture at a glance Both paths share the same back end: Unity Catalog enforcement and Databricks workspace identity. They differ in surface area and authentication model. Path A: connect to the four managed MCP servers This is the lightest setup. One mcp.json file, a Databricks Personal Access Token, and a shell-profile edit. In under 10 minutes Kiro is talking to Genie, SQL, Unity Catalog Functions, and Vector Search. Prerequisites A Databricks workspace on AWS with Unity Catalog enabled. A Databricks Personal Access Token (PAT) or OAuth token scoped to the MCP servers you plan to use ( sql , unity-catalog , genie , vector-search ). Unused PATs auto-revoke after 90 days. Kiro installed and launched at least once so ~/.kiro/ exists. Your workspace hostname in the format .cloud.databricks.com .

Generate a Databricks PAT In the Databricks workspace go to Settings, Developer, Access tokens, Manage, Generate new token. Set an expiration that aligns with your team's rotation policy. Select only the API scopes you need; least privilege beats the convenience of "everything." Copy the token immediately. Databricks does not display it again. Where Kiro stores MCP configuration Kiro reads MCP configuration from JSON in two scopes; workspace overrides user. User scope : ~/.kiro/settings/mcp.json applies to every workspace. Workspace scope : $PWD/.kiro/settings/mcp.json applies only to the current...