Gemini
Use Google's Gemini AI models for one-shot Q&A, text summarization, and content generation directly from the command line.
# Gemini Use Google's Gemini AI models for one-shot Q&A, text summarization, and content generation directly from the command line. The Gemini skill integrates Google's Gemini CLI into OpenClaw, enabling one-shot AI queries for question answering, text summarization, and content generation. It provides direct access to Gemini models without requiring an interactive session. ## How It Works The skill wraps the gemini CLI tool, allowing you to pass prompts as positional arguments for instant responses. It supports model selection, output format control (including JSON), and Gemini extensions for expanded capabilities. Authentication is handled through a one-time interactive login flow. ## Key Features - **One-Shot Queries**: Pass a prompt directly and get an immediate response without interactive mode - **Model Selection**: Choose specific Gemini models with the --model flag - **JSON Output**: Request structured JSON responses with --output-format json - **Extensions**: List and manage Gemini extensions for expanded functionality ## Requirements - **Gemini CLI**: The gemini command-line tool must be installed (available via Homebrew: brew install gemini-cli ) - **Google Account**: Authentication via Google account is required on first use ## Use Cases - **Quick Q&A**: Get instant answers to questions without leaving the terminal - **Text Summarization**: Summarize documents or text passages on the fly - **Content Generation**: Generate text, code, or structured data using Gemini models - **Data Extraction**: Use JSON output mode to extract structured information from unstructured text ## Installation Install via: `npx clawhub@latest install gemini`