Quickstart

Below is a quick guide to get you started with Vibe, for more details see Configuration to learn how to configure Vibe to your needs.

First, navigate to your project's root directory:

cd /path/to/your/project

cd /path/to/your/project

Once there, you can run Vibe:

vibe

vibe

This will start the interactive CLI interface.

If this is your first time running Vibe, it will:

• Create a default configuration file at ~/.vibe/config.toml
• Prompt you to pick a theme.

• Prompt you to enter your API key if it's not already configured, you can create one here.

• Save your API key to ~/.vibe/.env for future use

Alternatively, you can configure your API key separately using vibe --setup.

Once done, you can start interacting with the agent! Currently Vibe by default uses our Devstral 2 model, you can also use Devstral Small 2 or even your own custom model.

Learn how to use Vibe

We recommend using Vibe in interactive mode, but you can also use it in programmatic mode, learn more about both below.

Simply run vibe to enter the interactive chat loop, this mode simulates a chat interface with the agent.

• Multi-line Input: Press Ctrl+J or Shift+Enter for select terminals to insert a newline.
• File Paths: Reference files in your prompt using the @ symbol for smart autocompletion (e.g., > Read the file @src/agent.py).
• Shell Commands: Prefix any command with ! to execute it directly in your shell, bypassing the agent (e.g., > !ls -l).
• External Editor: Press Ctrl+G to edit your current input in an external editor.
• Tool Output Toggle: Press Ctrl+O to toggle the tool output view.
• Todo View Toggle: Press Ctrl+T to toggle the todo list view.
• Auto-Approve Toggle: Press Shift+Tab to toggle auto-approve mode on/off.

You can also directly start Vibe with a prompt with the following command:

vibe "Refactor the main function in cli/main.py to be more modular."

vibe "Refactor the main function in cli/main.py to be more modular."

Vibe also allows you to use slash commands and keyboard shortcuts for meta-actions and configuration changes during a session such as:

• /h, /help: Show a help message
• /config, /theme, /model: Edit configuration settings interactively
• /clear: Clear conversation history
• ...
• Ctrl+T Toggle todo view
• Shift+Tab Toggle auto-approve mode
• ... And more, use /help to display all available commands.

You can define your own slash commands through the skills system. Skills are reusable components that extend Vibe's functionality.

To create a custom slash command:

• Create a skill directory with a SKILL.md file
• Set user-invocable = true in the skill metadata
• Define the command logic in your skill

Example skill metadata:

---
name: my-skill
description: My custom skill with slash commands
user-invocable: true
---

---
name: my-skill
description: My custom skill with slash commands
user-invocable: true
---

Learn more about Agents.

Custom slash commands appear in the autocompletion menu alongside built-in commands.

You can define Agents and change them with Shift+Tab, below you can find a non-exaustive list of agents Vibe is packaged with:

• default: The standard default Vibe behavior, built for general usage requesting approval for tool execution.
• plan: Read-only agent for exploration and planning.
• auto approve: Automatically approves all tool executions without prompting.
• ...

Learn more about Agents.

You can run Vibe non-interactively by piping input or using the --prompt flag. This is useful for scripting, it allows you to run Vibe in a more streamline manner. This will not enter the interactive chat loop interface.

vibe --prompt "Refactor the main function in cli/main.py to be more modular."

vibe --prompt "Refactor the main function in cli/main.py to be more modular."

By default, it uses auto-approve mode.

When using --prompt, you can specify additional options:

• --max-turns N: Limit the maximum number of assistant turns. The session will stop after N turns.
• --max-price DOLLARS: Set a maximum cost limit in dollars. The session will be interrupted if the cost exceeds this limit.
• --enabled-tools TOOL: Enable specific tools. In programmatic mode, this disables all other tools. Can be specified multiple times. Supports exact names, glob patterns (e.g., bash*), or regex with re: prefix (e.g., re:^serena_.*$).
• --output FORMAT: Set the output format. Options:
  • text (default): Human-readable text output
  • json: All messages as JSON at the end
  • streaming: Newline-delimited JSON per message

Example:

vibe --prompt "Analyze the codebase" --max-turns 5 --max-price 1.0 --output json

vibe --prompt "Analyze the codebase" --max-turns 5 --max-price 1.0 --output json

Vibe is built to keep you in control. For substantial tasks (refactors, multi-file updates, cleanups), Vibe splits the work into steps it can execute safely and asks for confirmation before proceeding.
You can also cap the number of steps or set a maximum session cost, and control general permissions and usage in the config file, learn more about Configuration here.

Before editing files or running commands, Vibe shows a full preview and asks for confirmation.

After each step, you see exactly what was executed and the resulting output.

You’re ready to use Mistral Vibe in your terminal. Start with small tasks, try the interactive commands, and build up from there.