# databricks-solutions/ai-dev-kit

Databricks Toolkit for Coding Agents provided by Field Engineering

Repository: https://github.com/databricks-solutions/ai-dev-kit
Canonical: https://ross.abutalabs.com/products/ai-dev-kit
Language: Python
License: NOASSERTION
License Family: other
Topics: agents, claude, cursor, databricks, vibecoding
Last push: 2026-08-13T05:16:28+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 95, longevity 18
- inputs: {"age_days": 260, "days_push": 20, "days_rel": 36, "gap_med": 6.5, "n_releases_24m": 17}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1872, forks 410 (observed 2026-08-28T04:05:46.899804+00:00)

## What it is
A Databricks Field Engineering toolkit that provides skills, an MCP server, and a Builder App for AI coding agents working with the Databricks platform. Its skills have largely graduated into the official Databricks AI Tools distributed via the Databricks CLI, with this repo remaining as an installer, guide, and home for experimental tools.

## Use cases
- set up claude code or cursor to work with databricks
- install databricks skills for my coding agent
- run an mcp server that connects agents to databricks
- teach an ai agent about databricks asset bundles and genie
- bootstrap a vibecoding environment for lakebase and spark pipelines
- find experimental databricks developer tools from field engineering

## When to choose
- you use coding agents like claude or cursor against databricks and want curated skills and MCP integration
- you want guided onboarding for agent-driven databricks development
- you want to try experimental field-engineering tools before they graduate to official releases

## When to avoid
- you only need the official, engineering-supported skills - install Databricks AI Tools via the databricks CLI instead
- you need a fully supported MCP server with guaranteed maintenance - it is best-effort here
- you want a general-purpose agent framework unrelated to databricks

## Facets
- artifact type: plugin
- maturity: active
- function: mcp, agent-framework, developer-tools, cli, sdk
- domain: developer-tools, artificial-intelligence, large-language-models
- platform: python, cli, cross-platform
- tags: databricks, coding-agents, claude, cursor, vibecoding, skills, mcp-server, field-engineering, ai-agents, data-engineering

## Member repositories
- databricks-solutions/ai-dev-kit (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:46.899804+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:14:57.589661+00:00, confidence not recorded.
  - readme: https://github.com/databricks-solutions/ai-dev-kit (fetched 2026-08-28T04:05:46.899804+00:00, sha 9b454adf3835)
- Data as of 2026-08-30T08:39:29.467469+00:00.