Vallykrie/gemini-swarm-skill

5 stars · Last commit 2026-07-24

Fan out heavy-lifting tasks from Claude Code (or other AI coding agents) to parallel Gemini sessions via the Antigravity CLI, then integrate the results.

README preview

# gemini-swarm

[![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
[![GitHub stars](https://img.shields.io/github/stars/Vallykrie/gemini-swarm-skill?style=social)](https://github.com/Vallykrie/gemini-swarm-skill/stargazers)
[![GitHub last commit](https://img.shields.io/github/last-commit/Vallykrie/gemini-swarm-skill)](https://github.com/Vallykrie/gemini-swarm-skill/commits/main)

> **Gemini does the writing. Your orchestrator plans and reviews.**

gemini-swarm is a Claude Code plugin/skill (portable to other harnesses) that makes Gemini the *default* writer for substantial work. Whenever a task means writing more than a trivial amount of code or content, your orchestrator splits it into subtasks, fans each one out to a live `agy` ([Antigravity CLI](https://antigravity.google)) session running the right Gemini model, and integrates the results — all while its own context stays clean. It also ships **gemini-imagegen** for generating real images through Gemini's `generate_image` tool, something Claude can't do on its own.

<!-- Demo: docs/demo.gif or an asciinema cast showing `/gemini-swarm auto` fanning out and returning a run log. -->

## Contents

**New here?**
- [What it does](#what-it-does) — the idea, and the four steps it runs
- [Why this exists](#why-this-exists) — the unused-free-Gemini-Pro problem it solves
- [Use cases](#use-cases-when-it-pays-off) — is this the right fit for my task?

**Installing?**

View full repository on GitHub →