park-peter

Airflow To Dabs — Productivity skill for Claude Code

Productivity community

Coding agent skill that converts Apache Airflow DAGs to Databricks Declarative Automation Bundles (DABs).

How to install Airflow To Dabs

This entry records only its repository, not the path inside it, so there is no exact command to give. Open park-peter/airflow-to-dabs and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Airflow To Dabs does

Coding agent skill that converts Apache Airflow DAGs to Databricks Declarative Automation Bundles (DABs). Maps 40+ operators to Lakeflow Jobs task types, generates databricks.yml + job YAML + source files. Supports Hadoop/HDFS migration. Tested on Cursor, Claude Code, Codex, and Copilot.

Alternatives in Productivity

Workflow BundlesREADME — Workflow Bundles 45.5k ★
Getting Help — career-ops is an open source project maintained in limited time 32.5k ★
Om Correct — Sweep a corrected fact through the vault — finds every note restating it, applies the correction at the single 4.6k ★

README

airflow-to-dabs

A coding agent skill that converts Apache Airflow DAGs into [Databricks Declarative Automation Bundles](https://docs.databricks.com/en/dev-tools/bundles/) projects (formerly Databricks Asset Bundles; DABs).

Given an Airflow DAG file, the agent produces a complete bundle project — `databricks.yml`, `resources/*.yml` job definitions, and extracted `src/` source files — ready for `databricks bundle deploy`.

Platform Support

`SKILL.md` follows the open [Agent Skills](https://agentskills.io) format, so one skill folder works across coding agents:

Agent	Global	Project
Claude Code	`~/.claude/skills/airflow-to-dabs/` or the Claude Code plugin	`.claude/skills/airflow-to-dabs/`
Codex	`~/.agents/skills/airflow-to-dabs/`	`.agents/skills/airflow-to-dabs/`
Cursor	`~/.agents/skills/airflow-to-dabs/`	`.agents/skills/airflow-to-dabs/`
VS Code + GitHub Copilot	`~/.agents/skills/airflow-to-dabs/`	`.agents/skills/airflow-to-dabs/`
Gemini CLI, Windsurf, Kiro, Junie, Roo Code, OpenCode, and more	via `npx skills`	via `npx skills`

What It Does

Parses Airflow DAG files to extract tasks, dependencies, operators, schedules, and parameters
Maps 40+ Airflow operator types (including all Databricks provider operators) to DABs task type equivalents using a tiered mapping system
Source-aware routing: maps by connection, not class alone (operator → connection → intent → direction → destination → strategy) — Databricks SQL → sql_task, remote federatable DB → Lakehouse Federation, recurring ingestion → Lakeflow Connect; fail-closed on unresolved connections
Lakeflow Connect ingestion: routes recurring source→Delta ingestion (CDC, query-based, and foreign-catalog incl. Snowflake→Delta) to a DABs managed-ingestion pipeline