Build fluency in the terminology behind Google's terminal-based coding agent.
0 / 5 completed
1 / 5
At standup, a dev wants a terminal-based agent that can read and edit a local repo using Google's Gemini models. Which tool fits?
Gemini CLI is Google's open-source terminal agent that lets Gemini models read, edit, and run commands against a local codebase interactively. It brings agentic coding workflows to the command line rather than a browser IDE. This mirrors similar terminal-first coding agents from other model providers.
2 / 5
During a design review, the team wants Gemini CLI to only propose edits and ask before running any shell command. Which setting controls this?
Gemini CLI exposes an approval mode that determines whether shell commands and file writes execute automatically or require explicit confirmation first. Stricter settings reduce the risk of unintended changes on unfamiliar code. This is a standard safety control across terminal coding agents.
3 / 5
In a code review, a dev notices Gemini CLI pulled in project-specific instructions from a checked-in file automatically. What is this file's purpose?
Gemini CLI supports a checked-in context file that supplies persistent project-specific instructions, conventions, and constraints so the agent doesn't need them re-explained every session. This keeps agent behavior consistent across a team. It mirrors similar convention files used by other coding agents.
4 / 5
An incident report shows Gemini CLI ran a destructive command that hadn't been reviewed. What safeguard was likely misconfigured?
A destructive command running unreviewed usually points to an approval mode that granted broader autonomy than the task warranted. Tightening the mode for higher-risk operations, or scoping the working directory, mitigates this. Matching autonomy level to risk is a recurring theme across terminal agents.
5 / 5
During a PR review, a teammate asks how Gemini CLI differs from asking Gemini questions in a chat UI. What is the key distinction?
Unlike a chat interface that only returns text, Gemini CLI has direct execution access to read files, apply edits, and run commands in the local environment. This turns conversational suggestions into actual repository changes. That execution capability is what makes it an agent rather than just a chatbot.