playbooks

home / skills / openclaw / skills / gemini

gemini skill

/skills/steipete/gemini

This skill helps you use Gemini CLI for fast one-shot Q&A, summaries, and generation with simple prompts.

npx playbooks add skill openclaw/skills --skill gemini

Review the files below or copy the command above to add this skill to your agents.

Files (2)
SKILL.md
746 B
---
name: gemini
description: Gemini CLI for one-shot Q&A, summaries, and generation.
homepage: https://ai.google.dev/
metadata: {"clawdbot":{"emoji":"♊️","requires":{"bins":["gemini"]},"install":[{"id":"brew","kind":"brew","formula":"gemini-cli","bins":["gemini"],"label":"Install Gemini CLI (brew)"}]}}
---

# Gemini CLI

Use Gemini in one-shot mode with a positional prompt (avoid interactive mode).

Quick start
- `gemini "Answer this question..."`
- `gemini --model <name> "Prompt..."`
- `gemini --output-format json "Return JSON"`

Extensions
- List: `gemini --list-extensions`
- Manage: `gemini extensions <command>`

Notes
- If auth is required, run `gemini` once interactively and follow the login flow.
- Avoid `--yolo` for safety.

Overview

This skill provides a compact Gemini CLI for one-shot Q&A, summaries, and text generation. It is designed to run from the command line with a simple positional prompt and supports model selection, structured output formats, and extension management. The tool emphasizes non-interactive usage for reproducible results and archival convenience.

How this skill works

Run the CLI with a single prompt argument to produce one-shot answers or generations. You can choose a model with --model, request structured output like JSON with --output-format, and list or manage extensions with built-in subcommands. If authentication is required, the first run can be interactive to complete login before switching to non-interactive mode.

When to use it

  • Quick single-question Q&A from scripts or shell aliases
  • Generate summaries or short content without interactive sessions
  • Produce machine-readable output (JSON) for downstream tooling
  • Batch runs or cron jobs that call a CLI with different prompts
  • Explore or manage Gemini extensions from the terminal

Best practices

  • Provide the prompt as a positional argument to avoid interactive mode
  • Specify --output-format json when you need structured results
  • Use --model to control cost, latency, and behavior per task
  • Run once interactively only if authentication is required, then automate
  • Avoid the --yolo flag to maintain safety and predictable outputs

Example use cases

  • Call gemini "Summarize the latest meeting notes" in a CI step to archive summaries
  • Use gemini --model small "Answer: Is this code thread-safe?" from a pre-commit hook
  • Generate JSON with gemini --output-format json "Extract key metrics from this report" for ETL pipelines
  • List available extensions with gemini --list-extensions when managing plugins
  • Run gemini extensions install <name> to add functionality before batch processing

FAQ

Do I need to login to use the CLI?

Some configurations require authentication; run gemini once interactively and follow the login flow, then use non-interactive mode.

How do I get machine-readable output?

Add --output-format json to return JSON that can be parsed by your scripts or tools.