Google Challenges OpenAI with Gemini CLI Extensions
Right after OpenAI launched ChatGPT Apps, Google has entered the ring with its own ecosystem: Gemini CLI Extensions. A new race is on to become the definitive operating system for AI tools, allowing developers to integrate external services directly into their command line.
Gemini CLI is an open-source AI agent that operates from the terminal. The new extension system allows seamless integration with popular services like Dynatrace, Elastic, Figma, Harness, Postman, Shopify, Snyk, and Stripe, alongside Google's own tools such as Cloud Run, GKE, and Firebase.
How It Works
Each extension comes with a built-in playbook, which instructs the AI on how to use the integrated tool. This eliminates the need for manual setup; you can install an extension with a single command:
gemini extensions install <GitHub URL or local path>- Playbook-Driven: Each extension's 'playbook' guides the AI, ensuring seamless and predictable tool interaction without extra configuration.
- Standardized Protocol: The system is built on the Model Context Protocol (MCP) for consistent and reliable communication between the AI and the tools.
- Rich Integrations: The framework supports Google's own developer tools and even includes creative additions like 'Nano Banana' for image generation.
A Growing Ecosystem
To support this initiative, Google has launched the Gemini CLI Extensions Page, a public directory where developers can discover and install extensions from the community, official partners, and Google itself.