Set up the CLI

Prerequisites

If you haven't installed yet, run the install command from the installation page and complete browser authentication. Then come back here.

Get to know the CLI

Mind CLI has a built-in guide agent that knows everything about the CLI — commands, tools, settings, workflows, and best practices. Instead of reading docs, you can just ask it.

After launching a session, type:

Then have a conversation:

Press Shift+Tab to return to your main agent when you're ready to start working.

Start your first session

Navigate to any project directory and launch Mind:

The terminal UI opens — a rich chat interface with syntax highlighting, interactive panels, and visual tool progress.

Have a conversation

Type naturally. Mind understands your project context and can take action:

Core interactions

Reference files with @

Point Mind at specific files using @ with tab completion:

Plan before acting

Press Shift+Tab to switch to Plan mode. Mind outlines what it intends to do before making any changes — useful when you want to review the approach first.

Run shell commands directly

Prefix with ! to run commands without going through the agent:

Approve or reject changes

When Mind wants to run a command or write a file, you'll see an approval prompt. Press y to approve, n to reject, or Tab to drill into options and provide feedback.

Make Mind understand your project

Out of the box, Mind reads your files and infers context. But you can make it significantly smarter by telling it about your project's conventions. Create a steering file:

See Steering for the full guide on writing effective steering files.

Essential slash commands

Work with multiple agents

Mind ships with built-in agents for different tasks, and you can create your own:

• Default agent — general-purpose coding assistant
• Plan agent — breaks down complex tasks into structured plans (Shift+Tab)
• Guide agent — answers questions about Mind CLI itself (/guide)

For larger tasks, Mind can spawn subagents that work in parallel with isolated context. Press Ctrl+G to open the fleet monitor and watch them work in real time.

Stay in control

Beyond the per-action approval flow, you have fine-grained control over what Mind can do:

• Trust patterns — approve a command once, or trust a pattern (e.g., all npm run commands) so you're not interrupted for routine operations
• Tool permissions — use /tools trust <name> to pre-approve specific tools, or /tools untrust <name> to require explicit permission
• Reset anytime — /tools reset clears all runtime permissions

See Permissions for the full trust model.

Extend with external tools

Mind supports the Model Context Protocol (MCP), which lets it interact with databases, APIs, GitHub, and other services directly. Once you configure an MCP server, you can ask things like “list my open pull requests” or “query the staging database.”

See MCP integration to connect your first server.

Tips for your first session

• Press Shift+Enter to type multi-line prompts
• Press Shift+Tab to switch to Plan mode (agent plans before acting)
• Press Esc to cancel the agent or close any panel
• Type /help to see all available commands

Proxy configuration

Mind CLI respects standard proxy environment variables. Set these in your shell:

Debugging and troubleshooting

If you're having a problem with Mind CLI, run mind doctor to identify and fix common issues.