Skip to content

Gemini CLI Assistant

Lets an agent run one-shot questions, summaries and generation through the Gemini CLI.

Stars
11,367
Licence
MIT
Files
1 file

What it does

This skill teaches an AI agent to use Google's Gemini command-line tool for quick, single-question tasks such as answering a question, summarising text, or generating content. Once installed, the agent sends a prompt to Gemini from the terminal, can choose a specific model, and can ask for JSON output. It also knows how to list and manage Gemini CLI extensions. The skill itself is only instructions; it does not add new code.

Great for

  • Developers who already have the Gemini CLI installed and want their agent to call it for quick answers
  • People who want a model-generated summary or draft without leaving the agent workflow
  • Users who need structured JSON output from Gemini inside an automated script

Skip it if

  • Users who have no Google account or API access to Gemini, since the tool cannot run without authentication (the skill does not explain how to set up billing or keys)
  • Sensitive or confidential documents, because prompts are sent to Google's Gemini service over the network
  • Long interactive chat sessions, since the skill specifically tells the agent to avoid interactive mode

Try asking

  • “Ask Gemini a quick question about a programming error and return a short answer”
  • “Summarise this text file using Gemini and return the result as JSON”
  • “Generate a first draft of a product description with the gemini-pro model”
  • “List the Gemini CLI extensions I have installed”
Step-by-step guide: how to use Gemini CLI AssistantInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    gemini.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip gemini.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip gemini.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Keep prompts self-contained, because each call is one-shot and does not remember earlier turns
  • Do not use the --yolo flag, which the skill itself warns against; it bypasses safety confirmations
  • Run gemini once by hand first to complete login before asking the agent to use it

What's inside

  • SKILL.md

Redistributed unmodified from spinabot/brigade under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.