Skip to main content
skill CX Codex

databricks-iceberg — Claude Code skill: Iceberg tables on Databricks

databricks-iceberg — a skill for creating and managing Apache Iceberg tables on Databricks, syncing Delta with Iceberg via UniForm, and exposing tables to external engines through REST Catalog.

no license 327 installs ★ 408 updated 2026-10-24

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add databricks/databricks-agent-skills --skill databricks-iceberg -a claude-code
Open on GitHub

What databricks-iceberg does

The skill teaches the agent three core patterns: managed Iceberg tables that read and write as Iceberg everywhere; Delta tables with UniForm, which write as Delta in Databricks but read as Iceberg externally; and the Iceberg REST Catalog, which exposes UC tables to external engines (PyIceberg, Spark, Snowflake, Flink) over REST.

It embeds critical rules that prevent data corruption: do not install Iceberg libraries into DBR (version conflicts), do not manually manage metadata paths, choose between PARTITIONED BY and CLUSTER BY correctly, disable deletion vectors on v2 with clustering. The skill teaches which pattern fits each use case and how to set up external engine access with credential vending.

Requires Unity Catalog and Databricks CLI (≥ v1.0.0). Compatible across all listed platforms, but primary commands are SQL DDL/DML.

What it helps with

  • create an Iceberg table with correct partitioning for external engine access
  • enable reading Delta tables as Iceberg without rewriting data
  • configure REST Catalog so PyIceberg or Spark can read UC tables
  • integrate Snowflake with Databricks through Iceberg catalog
  • prevent version conflicts and metadata corruption when working with Iceberg

Example prompts for the agent

  • Create a managed Iceberg table from raw_events, partitioned by event_date
  • Enable External Iceberg Reads on my existing customers Delta table via UniForm
  • Set up external access to my tables through the Iceberg REST Catalog for a PyIceberg client

How to install databricks-iceberg

1Skills CLI (Vercel)

npx skills add databricks/databricks-agent-skills --skill databricks-iceberg -a claude-code

2Manually: put the files in place

~/.claude/skills/databricks-iceberg/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/databricks-iceberg/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about databricks-iceberg

How do I install databricks-iceberg in Claude Code?

Run npx skills add databricks/databricks-agent-skills --skill databricks-iceberg -a claude-code, then check ~/.claude/skills/databricks-iceberg/SKILL.md to verify.

What does databricks-iceberg do?

It equips the agent with knowledge of creating and managing Iceberg tables on Databricks, syncing Delta and Iceberg via UniForm, and granting external engines access to tables through REST Catalog.

Do I need to install an Iceberg library into Databricks Runtime?

No — DBR includes built-in Iceberg support. Adding a library will cause version conflicts and errors.

Similar skills

All skills for Data and databases →

From the same set30

The whole set →
databricks-ml-training databricks-ml-training is a skill for training classifiers and regressors on Databricks with XGBoost, scikit-learn, LightGBM, or PyTorch, and tracking runs with MLflow. CX Codex +1 354installs databricks-lakeflow-connect 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. CX Codex +1 348installs databricks-zerobus-ingest The databricks-zerobus-ingest skill helps route events, logs, CDC, and telemetry into Databricks Delta tables. It fits streaming setups that do not need a separate broker. CX Codex +1 331installs databricks-unstructured-pdf-generation databricks-unstructured-pdf-generation creates synthetic PDF files and paired evaluation questions locally, then copies them to Databricks Unity Catalog volumes for RAG testing. CX Codex +1 328installs databricks-synthetic-data-gen databricks-synthetic-data-gen creates synthetic Databricks data with Spark and Faker in Parquet, JSON, CSV, or Delta. Use it for test data, demos, and analysis-ready datasets. CX Codex +1 314installs databricks-genie-agents Builds and configures Databricks Genie Agents over Unity Catalog tables, metric views, and volume documents with SQL query routing. CX Codex +1 282installs databricks-setup-local The databricks-setup-local skill previews, configures, or troubleshoots a local uv-managed Python .venv for Databricks. Use it for Databricks Connect, compute targets, job tasks, or bundles. CX Codex +1 153installs databricks Gives your agent Databricks know-how: apps, jobs, pipelines, Lakebase, Model Serving and bundles. Installs into Claude. ★ 408repo stars

More in the catalog

Project sponsors

Companies that keep this analytics open