databricks-lakeflow-connect — Claude Code skill: setup
databricks-lakeflow-connect — a skill for Claude Code builds and debugs Lakeflow Connect ingestion from SQL Server, PostgreSQL, MySQL, and Oracle into Unity Catalog. It covers CDC and query-based loads.
no license 348 installs ★ 408 updated 2026-11-10
npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a claude-codeWhat databricks-lakeflow-connect does
databricks-lakeflow-connect runs in Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. It guides an agent through creating, running, and troubleshooting Lakeflow Connect pipelines that move data from SQL Server, PostgreSQL, MySQL, and Oracle into Delta tables in Unity Catalog.
The skill separates integrated CDC, standard CDC through a gateway, and query-based ingestion. It maps each mode to its pipeline fields, source tables, destinations, and compute checks. The databricks CLI version >= v1.2.0 is required, so the skill does not replace CLI setup.
What it helps with
- create a Lakeflow Connect pipeline for Unity Catalog
- choose CDC or query-based ingestion for a database
- map source tables to Unity Catalog destinations
- configure a gateway for standard CDC
- debug an ingestion pipeline failure
Example prompts for the agent
- Set up a Lakeflow Connect pipeline that moves PostgreSQL tables into Unity Catalog with CDC.
- Create a MySQL ingestion pipeline for Delta tables and choose the correct Lakeflow Connect mode.
- Investigate the pipeline error and compare its
serverlesssetting with other pipelines in the workspace.
How to install databricks-lakeflow-connect
1Skills CLI (Vercel)
npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a claude-code2Manually: put the files in place
~/.claude/skills/databricks-lakeflow-connect/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/databricks-lakeflow-connect/
1Skills CLI (Vercel)
npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a codex2Manually: put the files in place
.agents/skills/databricks-lakeflow-connect/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a cursor2Manually: put the files in place
.cursor/skills/databricks-lakeflow-connect/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install databricks/databricks-agent-skills skills/databricks-lakeflow-connect2Manually: put the files in place
.github/skills/databricks-lakeflow-connect/SKILL.md1Skills CLI (Vercel)
npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a gemini-cli2Platform CLI
gemini skills install https://github.com/databricks/databricks-agent-skills --path skills/databricks-lakeflow-connectManually: put the files in place
.devin/skills/databricks-lakeflow-connect/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about databricks-lakeflow-connect
How do I install databricks-lakeflow-connect in Claude Code?
Run npx skills add databricks/databricks-agent-skills --skill databricks-lakeflow-connect -a claude-code, then check ~/.claude/skills/databricks-lakeflow-connect/SKILL.md.
What does databricks-lakeflow-connect do?
It helps an agent create, run, and troubleshoot Lakeflow Connect ingestion from SQL Server, PostgreSQL, MySQL, and Oracle into Unity Catalog. The skill covers CDC, gateway-based CDC, and query-based ingestion.
What is required to use the skill?
The databricks CLI version >= v1.2.0 is required. Pipeline checks also use databricks pipelines list-pipelines and databricks pipelines get-update <id> <update-id> -o json.