All Skills
Automation & WorkflowClaude Plugins (Official)

oracle ai data platform workbench databricks migrator — Agent Skill

by oracle-samples · Updated

Drive the Oracle AI Data Platform (AIDP) Databricks Migration Toolkit in natural language. Plans and executes automated Databricks → AIDP migrations of notebooks, jobs, schedules, and catalog DDL — Pass-1 dependency rewrite + Pass-2 cell-by-cell execute/verify/fix on a live AIDP cluster with Claude.

At a glance

oracle-ai-data-platform-workbench-databricks-migrator is an Agent Skill by oracle-samples in the Automation & Workflow category. In Claude Code it installs with `/plugin install oracle-ai-data-platform-workbench-databricks-migrator@claude-plugins-official` after adding the anthropics/claude-plugins-official marketplace; other assistants use its SKILL.md instructions as rules. Its source repository has 57 GitHub stars (as of Oct 28, 2026).

Source 57GitHub stars

How to install oracle ai data platform workbench databricks migrator in your AI client

Setup for Claude Code, Claude Desktop, Cursor, Codex, Windsurf, Cline and VS Code — copy-paste ready.

How to install oracle ai data platform workbench databricks migrator in Claude CodeAnthropic's agentic CLI

  1. Add the marketplace

    claude
    /plugin marketplace add anthropics/claude-plugins-official

  2. Install the skill

    claude
    /plugin install oracle-ai-data-platform-workbench-databricks-migrator@claude-plugins-official

  3. Use it

    These are slash commands for a Claude Code session, not your shell. Claude auto-invokes the skill when your request matches its description — or type its slash command if it exposes one.

Documentation

Full README

Oracle AI Data Platform — Migrator for Databricks (Claude Code plugin)

Migrate Databricks notebooks, jobs, and catalogs onto Oracle AIDP — in natural language. Drives the *AIDP Databricks Migration Toolkit* (a Claude-with-tool-use migrator that runs cells live on the AIDP cluster, verifies, and self-fixes) end-to-end from Claude Code.

Canonical home: `oracle-samples/oracle-aidp-samples/ai/claude-code-plugins/oracle-ai-data-platform-workbench-databricks-migrator`. End users install via Anthropic's community marketplace (see Install).

This plugin is self-contained — it ships the full Python migrator engine bundled under engine/. After install, skills invoke the engine via ${CLAUDE_PLUGIN_ROOT}/engine/scripts/... — no separate clone required.

Status: v0.2.0 — self-contained release (engine bundled).

What it does

| Area | Skills | |---|---| | Bootstrap & setup | aidp-migrator-overview, aidp-migrator-bootstrap | | Plan & inventory | aidp-build-dag, aidp-check-data | | Notebook migration | aidp-migrate-job, aidp-fixup-cell, aidp-resume-migration | | Catalog migration | aidp-migrate-catalog, aidp-bucket-mapping | | Verification & quality | aidp-acceptance-contract |

Each skill is a single-file Markdown SKILL.md with a clear "When to use" trigger so Claude routes correctly without external direction.

What gets automated end-to-end

text
Databricks workspace                                    AIDP DataLake
─────────────────────                                   ──────────────
Unity Catalog / HMS schemas + tables (DDL)   ──┐
   │                                           ├──→  Catalog migration (batched DDL replay)
External s3:// table locations             ──┘            18 rewrite rules:
                                                          • 3-part → 2-part name flatten
                                                          • s3://→oci:// via bucket-map
                                                          • source format preserved (Delta stays Delta)
                                                          • delta.* / spark.sql.* catch-all scrub
                                                          • MV / streaming rejection
                                                          • CREATE SCHEMA COMMENT-colon strip

Notebooks (.dbc + .ipynb + .py)            ──→   Notebook migration (per workflow / per task)
Jobs + schedules + task DAGs                     Pass 1: %run dep tree, code-only rewrites
                                                 Pass 2: cell-by-cell execute on AIDP cluster,
                                                         4-way verify (status / stderr / Spark
                                                         logs / Opus eval), up to 10 fix attempts,
                                                         fixup_cell rewind for replays.
                                                 Output: per-job .ipynb + JOB_REPORT.md

Signature differentiators

  • Cell-by-cell execute / verify / fix loop — the migrator runs each Databricks-rewritten cell on a live AIDP cluster, parses the output, and self-corrects via Claude with tool use (14 tools: explore_path, suggest_oci_path, search_catalog, run_on_cluster, describe_table, list_schemas_and_tables, read_notebook_source, inspect_package_source, summarize_notebook, submit_code, make_note, get_cell_history, get_history_entry, fixup_cell).
  • Write-redirect sandbox schema — source data is never touched during migration. Every .saveAsTable(...) / INSERT INTO is silently redirected to a sandbox schema, then verified post-run.
  • Acceptance contract — for batch / streaming pipelines, declare PASS only after K consecutive empty-pending windows (consecutive-zero convergence).

Excerpt — the full document is on GitHub.

More Automation & Workflow skills

Catalog data refreshed . Something wrong? Report it.