On June 25, Google briefly unveiled its Gemini CLI, a new open-source project, via a blog post before quickly taking it down. The post described the free command-line tool that integrates Google’s Gemini AI models into developer terminals. The discovery by developer Meet Patel on X (formerly Twitter), with an archive.org link, allowed a view of the tool before it was removed. The Gemini CLI provides access to the Gemini 2.5 Pro model, with a 1 million token context. Key features include 60 model requests per minute and 1,000 per day for free with a Google login, integration with Gemini Code Assist for VS Code users, real-time web grounding via Google Search, and support for extensions, scripting, automation, and system prompts. Gemini CLI aims to simplify developer workflows. It provides an easy way to interact with AI for coding, debugging, and experimenting, as well as for generating videos and research. The tool’s open-source licensing under Apache 2.0 allows developers to inspect the code, build extensions, and create custom workflows. The reasons behind the blog post’s removal are uncertain. The fact that the post was created by Google engineers Taylor Mullen and Ryan Salva suggests that it was not a hoax. Although the official launch is postponed, the cached content allows for a comprehensive look at the Gemini CLI and the future of AI tools.
