openai-agents-gemini

Run OpenAI Agents on Gemini via the OpenAI-compatible API.

Updated Jan 6, 2026
One-click install
npx skills add https://github.com/Ahmednoorani258/Hackathon2-phase3 --skill openai-agents-gemini
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openai-agents-gemini
Source: https://github.com/Ahmednoorani258/Hackathon2-phase3/tree/main/.claude/skills/openai-agents-gemini
Command: npx skills add https://github.com/Ahmednoorani258/Hackathon2-phase3 --skill openai-agents-gemini

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai-agents, python-dotenv, and includes assets (resource) and references (resource) components.

What problem does it solve?

Configures an OpenAI Agents SDK setup to run with Google's Gemini backend via the OpenAI-compatible API, enabling developers to build Gemini-backed agents quickly.

Core Features & Use Cases

  • OpenAI Agents SDK integration wired to Gemini, using AsyncOpenAI client and OpenAIChatCompletionsModel.
  • Tool definitions with @function_tool for weather and travel planning.
  • A runnable example script (assets/agent_gemini.py) to test and iterate locally.
  • Environment template guidance for GEMINI_API_KEY and OPENAI_API_KEY.

Quick Start

Run the Gemini-enabled agents locally by executing the provided asset script to observe tool usage and responses.

Frequently Asked Questions about openai-agents-gemini

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

FAQPage Schema
How do I run OpenAI Agents with Gemini backend?

To run OpenAI Agents with Gemini backend, this Skill uses an AsyncOpenAI client wired to the Gemini base_url and an OpenAIChatCompletionsModel wrapper. It provides a runnable Python asset script demonstrating end-to-end execution.

Can I use tool calling with OpenAI Agents SDK on Gemini?

Yes, you can implement tool calling with OpenAI Agents SDK on Gemini. The Skill includes predefined tools using the @function_tool decorator for weather and travel planning, supporting tool-calling workflows via the OpenAI-compatible API.

Do I need an OpenAI API key to use the Gemini backend for agents?

You need a GEMINI_API_KEY for the Gemini backend, and the environment template also includes guidance for an OPENAI_API_KEY. These keys configure the AsyncOpenAI client to route requests properly.

What is the best way to test Gemini-backed agents locally?

The best way to test Gemini-backed agents locally is by executing the provided asset script, assets/agent_gemini.py. This ready-to-run Python example allows you to observe tool usage and responses directly.

Does running OpenAI Agents on Gemini require specific Python dependencies?

Running OpenAI Agents on Gemini requires the openai-agents and python-dotenv dependencies. These packages facilitate the SDK integration and environment variable loading for the AsyncOpenAI client configuration.

Why use OpenAIChatCompletionsModel for Gemini-backed agents?

Using OpenAIChatCompletionsModel for Gemini-backed agents bridges the OpenAI Agents SDK with Google's Gemini via the OpenAI-compatible API. This model wrapper enables seamless integration and local testing for Gemini-backed agent workflows.