This skill sends a question to ChatGPT and Gemini, then brings back what each of them would do. It helps software engineers, architects, and automation builders make implementation decisions with more than one AI perspective in hand. The workflow queries external LLM APIs, analyzes their suggestions, and produces a synthesized plan that notes where the models agreed, diverged, or raised concerns.
Follow the documented setup, then try a first task.
Installs: Complete `llm-council.skill` bundle · Claude
Run from: The directory where Claude is run
In Claude, upload the `llm-council.skill` file to install the skill. This is a manual action, not a terminal command.
Obtain an OpenAI API key from the OpenAI Platform and a Gemini API key from Google AI Studio.
Create `.env` in the directory where Claude is run with these lines: OPENAI_API_KEY=sk-your-openai-key-here GEMINI_API_KEY=your-gemini-key-here OPENAI_MODEL=gpt-5-nano-2025-08-07 GEMINI_MODEL=gemini-3-flash-preview Replace `sk-your-openai-key-here` and `your-gemini-key-here` with the corresponding API keys. Keep the file secure and do not commit it to version control.
Uses llm-council
Use the llm-council skill to consult the council: How should I architect a real-time data pipeline for IoT sensors?
Claude
Gemini CLI