LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
ToolRisingofficial publisher

run_shell_command, write_file, google_web_search (Gemini CLI)

Tool : A precise action the AI can call · requires gemini-cli

The built-in tools of Google's CLI, each with confirmation.

Correct tools, but the CLI is transitioning to Antigravity.

GitHub · 108,162 ★ Docs
Overall score
7.2 /10
Security: one caveat, see the card
Utility
6.0
Reliability
10.0
Security
One caveat
Ease
5 min · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Working with files, shell commands, and web searches from a terminal usually means switching between multiple tools. Gemini CLI brings these actions together in a single command-line agent powered by Google's Gemini model.

In practice, three built-in tools cover the essentials: run_shell_command executes commands in the terminal, write_file creates or modifies files, and google_web_search grounds responses in real-time search results. Each sensitive action requires explicit confirmation before it is carried out.

Gemini CLI is aimed at developers who live in the command line and want to automate repetitive tasks, analyse a code repository, or generate files without leaving the terminal. It can also run in non-interactive mode to integrate into scripts or CI workflows.

It is not suited to users who prefer a graphical interface, or to those who need a secure production environment without prior sandbox configuration: executing shell commands remains a risky operation that requires constant attention.

What it offers

  • ·run_shell_command executes shell commands in the terminal with a confirmation prompt before each action.
  • ·write_file creates or modifies local files, also subject to explicit confirmation.
  • ·google_web_search grounds model responses in real-time Google Search results.
  • ·One-million-token context window for analysing large codebases or long documents.
  • ·MCP (Model Context Protocol) support for connecting third-party tools and servers.
  • ·Non-interactive (headless) mode for integrating Gemini CLI into scripts or automated pipelines.
  • ·GEMINI.md context files to customise agent behaviour per project.
  • ·Session saving and resumption via the checkpointing system.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
cli
Host
gemini-cli
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Capabilities
readswritesexecutesnetwork
Your data
Data stays local
Maintenance
Active
Last release
20 October 2026

Links and repository checked on 25 October 2026 · added on 1 June 2025

How to install run_shell_command, write_file, google_web_search (Gemini CLI)

1. Make sure Node.js is installed on your machine (required for npm). 2. Install Gemini CLI globally: npm install -g @google/gemini-cli 3. Launch the agent with the command: gemini 4. On first launch, choose your authentication method: Google sign-in (OAuth) for the free tier, or a Gemini API key via the GEMINI_API_KEY environment variable. 5. Security note: before using run_shell_command, configure trusted folders and enable sandboxing if your environment supports it, to limit the scope of executed commands.
npm install -g @google/gemini-cli

Official documentation ↗

In the same category

Whole category Code →