On This Page8 sections
Editorial snapshot from September 2026. Prices, models and interface labels may change; this is a practical guide, not a live product feed.
Google’s open-source terminal agent for Gemini, project context and command-line automation.
Coding agent · Google · macOS · Windows · Linux
Suitable workflows
Developers using Google accounts, Gemini API keys or Vertex AI.
Limitations and billing
Quotas depend on authentication. Workspace accounts may need a Google Cloud project.
Capabilities we verified
Agent workflow · CLI surface · MCP · Skills · Rules / instructions · Hooks · Bring your own key
This is an evidence-based shortlist, not an exhaustive feature inventory. Omitted capabilities have not been checked for this record.
Pricing & usage
Free access; see limits
Account quota or metered Gemini / Vertex AI usage. The cost of metered usage or an external agent can be separate from the subscription.
Models & access
Google sign-in, Gemini API key, or Vertex AI credentials.
Verified families: Gemini.
Install and start the CLI
Install a supported Node.js version and npm; use the stable release for normal projects.
npm install -g @google/gemini-cliOpen your project folder in a terminal, then run:
geminiAccount and authentication
Choose Google sign-in, GEMINI_API_KEY, or Vertex AI on first launch.
Project configuration and tools
Use GEMINI.md for project context; configure extensions and MCP separately from authentication.
First task to try
Map the main modules and test commands before asking Gemini to change a file.
Script usage
This example asks for a project summary. Tool permissions still apply; a prompt is not a sandbox.
gemini -p "Summarize the project structure without modifying files."Related CLI tools
Qwen Code — Qwen’s open-source terminal agent for repository work and configurable model access.
Codex — OpenAI’s coding agent for local projects and cloud tasks.
Grok Build — Grok’s coding agent with a terminal interface, script mode and ACP integration.
Project evaluation checklist
- Open a small, familiar project and confirm that your essential language tooling works.
- Give the agent one bounded change with an observable acceptance criterion.
- Review the diff, run the project’s checks and compare the result with your current workflow.
- Inspect usage after the task. Estimate a month of similar work, including any separate model or agent bill.
Related TRAE guides
Continue Reading
More articles connected to the same themes, protocols, and tools.
Referenced Tools
Browse entries that are adjacent to the topics covered in this article.








