Skip to content

feat: Add databricks power#128

Closed
antonyprasad-db wants to merge 1 commit into
kirodotdev:mainfrom
antonyprasad-db:add-databricks-power
Closed

feat: Add databricks power#128
antonyprasad-db wants to merge 1 commit into
kirodotdev:mainfrom
antonyprasad-db:add-databricks-power

Conversation

@antonyprasad-db
Copy link
Copy Markdown

Summary

Adds the Databricks AI Dev Kit as a Kiro Power. Currently no Databricks power exists in this catalog — this fills that gap.

The Power provides:

  • 44 MCP tools (180+ operations) for executing Databricks operations directly: SQL, jobs, pipelines, Unity Catalog, Genie, Agent Bricks, model serving, dashboards, Lakebase, Vector Search, and more
  • 34+ expert skills for building data pipelines, ML workflows, BI dashboards, and AI/agentic applications on Databricks across AWS, Azure, and GCP
  • Complete onboarding flow with three authentication options (OAuth via CLI, existing ~/.databrickscfg profile, or PAT)

What's included

databricks/
├── POWER.md          # Manifest + step-by-step onboarding
├── mcp.json          # Wires databricks-mcp-server with credential profile support
└── steering/
    └── README.md     # Skill index — installer populates this on activation

Structure matches the existing aws-agentcore power.

How the Power works

On first activation, the agent runs the official Databricks AI Dev Kit installer in non-interactive Kiro mode:

bash <(curl -sL https://raw.githubusercontent.com/databricks-solutions/ai-dev-kit/main/install.sh) --tools kiro --global --silent

The installer downloads the MCP server, creates a venv, and pulls 34+ skills into ~/.kiro/skills/. The Power's onboarding flow then copies the skills into its own steering/ directory so they're scoped to Power activation (no global namespace pollution).

This means the Power stays lean — skills are always fetched from the latest ai-dev-kit repository rather than bundled and going stale.

Foundation

The Kiro IDE installer support that this Power depends on was merged in ai-dev-kit PR #511.

Contributors

  • @antonyprasad-db — Senior Specialist Solutions Architect, Databricks (AWS Partnership) — Kiro IDE installer foundation in ai-dev-kit PR #511; primary maintainer of this Power
  • @venkatavaradhanv — Solutions Architect, AWS — prototyped the catalog-ready Power packaging at venkatavaradhanv/databricks

Checklist

  • POWER.md follows the canonical frontmatter (name, displayName, description, keywords, author)
  • mcp.json wires a valid MCP server with safe defaults (disabled: true until user configures credentials)
  • steering/ directory present
  • No proprietary or internal-only references — all links are to public GitHub repos and public Databricks docs
  • Tested against the live ai-dev-kit installer flow

Adds the Databricks AI Dev Kit as a Kiro Power. Provides 44 MCP tools
(180+ operations) and expert skills for building data pipelines, ML
workflows, dashboards, jobs, and AI/agentic applications on the
Databricks platform across AWS, Azure, and GCP.

The Power's onboarding flow runs the official Databricks AI Dev Kit
installer (databricks-solutions/ai-dev-kit) in Kiro-targeted mode and
copies skills into the Power's steering directory.

- POWER.md: manifest + complete onboarding flow with auth options
- mcp.json: wires databricks-mcp-server with credential profile support
- steering/README.md: skill index pointing back to the ai-dev-kit repo

Authored by Antony Prasad Thevaraj (Databricks); catalog-ready
packaging prototyped by Venkatavaradhan Viswanathan (AWS).

Co-authored-by: Isaac
@github-actions
Copy link
Copy Markdown

Hi @antonyprasad-db, thank you for your contribution!

Please note that if you haven't already, you would also need to submit your power officially at kiro.dev/powers/submit so it can be reviewed for listing in the Kiro powers registry.

@antonyprasad-db antonyprasad-db marked this pull request as draft May 19, 2026 03:06
@antonyprasad-db
Copy link
Copy Markdown
Author

Marking this as draft — pausing for an internal review of Databricks brand language and standards (the POWER.md uses `author: "Databricks Solutions"` and the "Databricks AI Dev Kit" name; both should be cleared internally before this lands publicly in the catalog).

Will flip to ready-for-review once that internal review is complete. Apologies for the noise; the technical content is sound — this is purely a brand/standards check.

In the meantime, happy to take any directional feedback from maintainers (e.g., should we name the folder `databricks` or `databricks-ai-dev-kit`?).

@antonyprasad-db
Copy link
Copy Markdown
Author

Closing this temporarily for internal review on the Databricks side. The technical content is sound (it wraps the databricks-solutions/ai-dev-kit installer, which is the canonical toolkit), but a few items need to clear before this lands publicly under `author: "Databricks Solutions"`:

  1. Brand sign-off on the "Databricks AI Dev Kit" naming and "Databricks Solutions" authorship in an external public catalog
  2. License clarity on the upstream ai-dev-kit repo (currently shows `NOASSERTION` on GitHub — clearing this internally so derivative works have a defensible OSS basis)
  3. Architecture polish — bundle a small set of core skills directly in `steering/` so the Power has value standalone, simplify the dual-step skill copy pattern, add an activation smoke test

Will re-open as v2 once those clear (expected ~1 week). Apologies for the noise.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant