gemini-cli-extensions/dataproc
claude-codeOfficialSkills for Dataproc
Dataproc Agent Skills is an open-source extension by Google that lets AI coding agents like Claude Code and Codex manage Google Cloud Dataproc clusters and jobs through natural language. Skip the gcloud syntax and just describe what you need.
Our Review
Maintained by Google under the gemini-cli-extensions org and released under Apache 2.0, Dataproc Agent Skills connects your AI coding agent directly to Google Cloud's managed Apache Spark and Hadoop service. It turns cluster and job management into natural-language requests instead of memorized gcloud commands.
What Dataproc Agent Skills adds to your AI agent:
- •Natural-language cluster control ask your agent to list clusters in a region or pull full details for a named cluster, no gcloud syntax required.
- •Job monitoring in plain English check the status of a specific job ID or list every failed job in your project with a single prompt.
- •Full resource lifecycle visibility inspect the state of your Dataproc clusters and jobs across the project from inside your agent session.
- •Open Agent Skills format packaged in the open skill format used across the gemini-cli-extensions family, installable with the
npx skills addcommand. - •IAM-scoped access runs against your real Google Cloud project using Application Default Credentials and standard Dataproc Viewer or Editor roles.
- •Project and region config set
DATAPROC_PROJECTandDATAPROC_REGIONenvironment variables once, then work conversationally.
Compatibility:
- •Claude Code install via the plugin marketplace command (
/plugin marketplace add) on Claude Code v2.1.94 or higher. - •Codex clone the repo into the Codex plugins directory on Codex v0.117.0 or higher.
- •Antigravity works with Antigravity CLI v1.6.0+ or Antigravity 2.0 (v2.0.0+), with migration support from Gemini CLI extensions.
Limitations:
The project is in beta (pre-v1.0) and the maintainers warn of breaking changes before the first stable release, so pin to a release tag rather than tracking main. It is read and monitoring focused around listing clusters, checking jobs, and pulling details, not a full provisioning suite. It requires an existing Google Cloud project with the Dataproc API enabled, so it is useful only inside the Google Cloud ecosystem. Adoption is early, with 1 GitHub star as of June 2026, and support is community-driven through the project's feedback form.
Our Verdict
Dataproc Agent Skills is a focused, official bridge between AI coding agents and Google Cloud's managed Spark and Hadoop service. If you already run data workloads on Dataproc, it removes the gcloud lookup loop and lets you query cluster and job state by typing what you mean.
For Vibe Builders working near data infrastructure, this lowers the barrier to checking on a pipeline without learning the full gcloud command surface. The trade-off is that Dataproc itself is a technical platform, so you still need a configured Google Cloud project and the right IAM roles before the conversational layer helps.
For Developers and data engineers on GCP, the value is speed. Listing failed jobs, checking a job ID, or pulling cluster details becomes a one-line prompt inside Claude Code, Codex, or Antigravity, kept honest by real Dataproc Viewer or Editor permissions.
Skip it if you are not on Google Cloud, if you need full provisioning and job-submission automation today rather than monitoring, or if you cannot accept beta software that may introduce breaking changes before v1.0. For multi-cloud Spark management or a production control plane, wait for a stable release or use the native Dataproc console and gcloud directly.
Frequently Asked Questions
What is Dataproc Agent Skills?
Dataproc Agent Skills is an open-source extension maintained by Google in the gemini-cli-extensions organization. It gives AI coding agents a set of skills to manage Google Cloud Dataproc clusters and jobs using natural language instead of gcloud commands. You can list clusters, check job status, find failed jobs, and pull cluster details by simply describing what you want.
Is Dataproc Agent Skills free and open source?
Yes, Dataproc Agent Skills is free and released under the Apache 2.0 license, so you can use, modify, and self-host it without a fee. The skills themselves cost nothing, but they operate against your own Google Cloud project, so you pay Google's standard Dataproc usage charges for any clusters and jobs you run. As of June 2026 the project is in beta (pre-v1.0).
Which AI agents work with Dataproc Agent Skills?
Dataproc Agent Skills works with Claude Code (v2.1.94 or higher), Codex (v0.117.0 or higher), and Antigravity (CLI v1.6.0+ or Antigravity 2.0). Choose Claude Code when you want the simplest install through the plugin marketplace command. Choose Codex or Antigravity when those agents are already your primary coding environment and you prefer the clone-and-copy install flow.
What can Dataproc Agent Skills actually do?
Dataproc Agent Skills lets you manage the lifecycle of Google Cloud Dataproc resources through conversation. Common requests include listing all clusters in a region, getting details for a named cluster, checking the status of a specific job ID, and listing every failed job in your project. It uses your Dataproc Viewer or Editor IAM role, so its reach matches the permissions you grant it.
How do I install Dataproc Agent Skills in 2026?
First enable the Dataproc API on your Google Cloud project and set up Application Default Credentials, then set the DATAPROC_PROJECT and DATAPROC_REGION environment variables. In Claude Code, run the plugin marketplace add command pointing at the repo, then install the dataproc plugin. You can also install across agents with the open Agent Skills tool using the npx skills add command.
What is dataproc?
Dataproc Agent Skills is an open-source extension by Google that lets AI coding agents like Claude Code and Codex manage Google Cloud Dataproc clusters and jobs through natural language. Skip the gcloud syntax and just describe what you need.
What license does dataproc use?
dataproc uses the Apache-2.0 license.
What are alternatives to dataproc?
Explore related tools and alternatives on My AI Guide.
Open source: source code publicly visible
Anyone can inspect exactly what this repo does on GitHub before using it. Apache-2.0 licensed.
Reviewed by My AI Guide for relevance, quality, and active maintenance before listing.
Install in Claude Code:
/install dataprocTopics
