gemini
CLI interface for Gemini AI, enabling one-shot model inference, text generation, and JSON-formatted data extraction for OpenClaw users.
Introduction
The Gemini CLI is a specialized skill for the OpenClaw assistant ecosystem designed for direct, terminal-based interaction with Google's Gemini models. It provides a lightweight, efficient wrapper for users who need to perform one-shot Q&A, rapid content summarization, and data generation without the overhead of a full conversational chat interface. By leveraging the command line, this tool integrates seamlessly into developer workflows, automation scripts, and terminal-centric productivity environments.
-
Perform one-shot inference using positional prompts directly from your shell.
-
Support for custom model selection via the --model flag to toggle between different Gemini versions.
-
Built-in structured data output using --output-format json, ideal for pipeline integration and machine parsing.
-
Extension management system allowing users to list and control auxiliary Gemini-based plugins through the gateway interface.
-
Native compatibility with OpenClaw auth flows, requiring only a one-time interactive login to persist credentials securely.
-
Use this tool for quick CLI-based drafting, summarizing log outputs, or generating boilerplate code snippets.
-
Avoid interactive mode prompts; always pass your full query as a positional argument for optimal performance.
-
Use the --list-extensions command to discover and manage installed capability modules.
-
Safety warnings: The tool includes a --yolo flag; use caution as this may bypass standard safety filters in certain contexts.
-
Prerequisites: Ensure your OpenClaw gateway is correctly onboarded and authenticated before executing commands.
Repository Stats
- Stars
- 366,013
- Forks
- 75,040
- Open Issues
- 6,964
- Language
- TypeScript
- Default Branch
- main
- Sync Status
- Idle
- Last Synced
- Apr 29, 2026, 06:10 AM