gemini

What This Skill Does

The Gemini skill is a powerful command-line interface (CLI) for OpenClaw that enables seamless integration with Google's Gemini AI models. It acts as a versatile bridge for one-shot tasks, ranging from quick Q&A sessions to complex content generation and detailed data analysis. The skill supports both direct CLI interactions for concise outputs and a specialized file-runner mode for tasks involving large tokens, extensive documentation, or long-form creative writing. By automating the interaction with Gemini, it allows users to pipe AI-generated insights directly into their workflow, manage large reports, or perform structural analysis on complex video transcripts without leaving the terminal.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/chenbaiyujason/gemini-file

Once installed, ensure you have completed the authentication process by running gemini once manually in your terminal and following the interactive login prompts provided by the service.

Use Cases

This skill is designed for scenarios where efficiency and reliability are paramount:

• Creative Content Generation: Write essays, articles, or reports of significant length and save them directly as files for later editing.
• Complex Data Processing: Feed dense data or transcripts into the AI and receive structured outputs, such as JSON-formatted data for further automation tasks.
• Video Intelligence: Leverage the gemini-3-flash-preview model to parse long video transcripts or visual metadata, providing detailed, structured summaries.
• Technical Documentation: Automate the creation of documentation, code comments, or configuration files based on specific project requirements.

Example Prompts

1. "写一篇关于量子计算未来趋势的 1500 字深度报道，并保存为 quantum-future.txt"
2. "解析这段会议记录的文本，输出为 JSON 格式，并包含关键决策点"
3. "使用 gemini-3-flash-preview 模型总结这个长视频并生成详细的报告文件"

Tips & Limitations

• Safety First: Avoid using the --yolo flag to maintain system integrity and output safety.
• Performance: For large documents or video analysis, always prefer the file-runner mode over standard CLI prompts to avoid output truncation or terminal buffer limitations.
• File Management: Files are automatically generated in the outputs/ directory. If you require specific organizational structures, use the --output-dir argument to direct files to your preferred project folders.
• Environment: If the CLI fails to execute, check your local Python installation, as the file-runner utility requires an active Python 3 environment.