collaborating-with-gemini

Bridge Gemini CLI interactions to Codex for prototyping, debugging, and code review.

130|17|Updated Dec 20, 2025
One-click install
npx skills add https://github.com/appautomaton/agent-designer --skill collaborating-with-gemini-appautomaton
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: collaborating-with-gemini
Source: https://github.com/appautomaton/agent-designer/tree/main/.codex/skills/collaborating-with-gemini
Command: npx skills add https://github.com/appautomaton/agent-designer --skill collaborating-with-gemini-appautomaton

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gemini, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Bridges Codex workflows with Gemini CLI, enabling a dedicated collaborator for prototyping, debugging, and code review while maintaining Codex as the primary implementer and reducing token-heavy back-and-forth.

Core Features & Use Cases

  • Bridge-based collaboration: A Python-based bridge communicates with Gemini via a deterministic JSON stream, preserving session state and output format.
  • Multi-turn sessions: Reuses SESSION_ID across messages to maintain conversation context.
  • Safety and workflow: Enforces patch-first changes, discourages direct file edits, and logs collaboration state for auditability.
  • Use Case: Propose a minimal code patch or refactoring and obtain a patch that can be applied via a code review workflow.

Quick Start

Install the Gemini CLI and run the bridge with a starting prompt to begin a multi-turn coding session.

Frequently Asked Questions about collaborating-with-gemini

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use Gemini CLI for code collaboration without leaving Codex?

Gemini CLI for code collaboration is bridged to Codex through a Python script that communicates via a deterministic JSON stream, allowing you to prototype, debug, and review code while Codex remains the primary implementer.

Does the Gemini CLI bridge support multi-turn coding sessions?

Multi-turn coding sessions are supported by reusing a SESSION_ID across messages, which maintains conversation context and preserves session state throughout your prototyping and debugging workflow.

What is the best way to minimize token usage when bridging Gemini for code review?

To minimize token usage during code review, the Gemini bridge operates with file and line-based prompts, enforcing a patch-first workflow that generates structured patches rather than engaging in token-heavy back-and-forth.

Can I apply direct file edits from Gemini CLI when collaborating with Codex?

Direct file edits from Gemini CLI are discouraged; the bridge script enforces a patch-first workflow for code changes, ensuring safe output and logging collaboration state for auditability instead of modifying files directly.

Do I need the Gemini CLI installed to run the Codex collaboration bridge?

You need the Gemini CLI installed to run the collaboration bridge, which serves as the entry point for starting a multi-turn coding session with a prompt to generate patches for your code review workflow.