# Gemini CLI playbooks

Gemini CLI is Google's open-source agent for the terminal, released under the Apache 2.0 license. It reads and edits files, runs shell commands, searches the web and calls tools through MCP. Google moved most individual users to Antigravity CLI in June 2026, so it now serves API key and Gemini Code Assist business users.

- Vendor: Google
- Kind: coding-agent
- Surfaces: terminal, ide, api
- Connects: Add MCP servers in ~/.gemini/settings.json. Agent Skills, extensions, custom commands, hooks and subagents are documented features. Project context lives in GEMINI.md files.

- Docs: https://geminicli.com/docs/

## Good at

- Editing a codebase and running shell commands from the terminal
- Scripted runs with -p and JSON output for CI jobs
- Grounding answers with Google Search and web fetch tools
- Pull request reviews and issue triage through the run-gemini-cli GitHub Action

## Limits

- Google's blog says Gemini CLI stopped serving Google AI Pro, Ultra and free individual users on June 18, 2026. Those users must move to Antigravity CLI.

- It still works with paid Gemini API keys and Gemini Code Assist Standard or Enterprise licenses, so access needs billing or a business plan.
- It acts on your machine, so a vague prompt can change files you did not mean to change. Checkpointing and sandbox modes reduce the risk.

## Playbooks (5)

- [Build a personal knowledge base wiki that a coding agent maintains from your notes](https://agentshandbook.com/p/build-a-personal-knowledge-base-wiki-from-your-notes.md): You have a folder of interlinked markdown pages with an index and a change log, built from your sources, that you can read in Obsidian and query through the agent, with every change reviewable in git.
 (trust: corroborated, risk: low)
- [Map an unfamiliar codebase into a checked architecture document before you change it](https://agentshandbook.com/p/map-an-unfamiliar-codebase-before-you-change-it.md): You have a docs/architecture.md that names the entry points, modules and data flow with file references, a list of open questions, and a short instruction file with the build and test commands.
 (trust: corroborated, risk: low)
- [Sort a messy Downloads or photos folder without losing a single file](https://agentshandbook.com/p/sort-a-messy-downloads-folder.md): Your folder is sorted into a small set of named subfolders, nothing is deleted, and you have a log of every move that you can reverse. (trust: corroborated, risk: high)
- [Test your local web app in a real browser and fix console errors in one loop](https://agentshandbook.com/p/test-local-web-app-in-browser-and-fix-console-errors.md): You get a table of the console errors and failed requests on each route you named, a small fix for each one, and proof from a reload that the error is gone.
 (trust: corroborated, risk: low)
- [Upgrade a framework across a repo with a coding agent and the test suite as the gate](https://agentshandbook.com/p/upgrade-a-framework-across-a-repo-with-tests-as-the-gate.md): The repo runs on the new version on a branch of small commits, the full test suite passes with no tests removed, and you have a list of behaviour changes to check in staging.
 (trust: corroborated, risk: medium)

## Lessons

- [Desktop and terminal agents delete files while they tidy up](https://agentshandbook.com/lessons/agents-delete-files-while-organizing.md)
- [Coding agents write code for old library versions or invent APIs](https://agentshandbook.com/lessons/coding-agents-use-outdated-library-apis.md)