Skip to main content

​
What It Does

Monk lives in your IDE as a native extension. You chat with it, deploy infrastructure, monitor your app, and manage your entire system. You don’t leave your editor.

​
Getting Started

Download Monk for your IDE: https://monk.io/downloads Supported IDEs:
  • VS Code - Microsoft’s Visual Studio Code
  • Cursor - AI-first code editor
  • Antigravity - Google’s AI-first editor
  • Windsurf - AI-powered IDE
  • Other VS Code forks - Any VS Code-compatible editor works
Install the extension. You’re ready to deploy.

​
How It Works

​
Connected Context

Monk sits in your IDE and builds context by connecting three layers:
  1. Your codebase - Source code, dependencies, configuration files
  2. Runtime state - Running containers, service health, logs, metrics
  3. Infrastructure - Cloud resources, VMs, databases, networking
With all three connected, Monk can:
  • Deploy based on what’s in your code
  • Troubleshoot using live runtime data
  • Optimize infrastructure based on actual usage
  • Give you accurate cost numbers from real resource utilization
Example: You ask “Why is the API slow?” Monk examines your code, checks live container CPU/memory, reviews database query performance, and identifies the bottleneck. One chat interface. One answer.
Code privacy: Your source code is analyzed to understand deployment needs but is not stored and not used for training. See Security for details on how Monk interacts with your codebase.

​
Monk Chat Window

Monk has its own dedicated chat interface inside your IDE. What you can do:
  • Deploy applications: deploy this project
  • Check status: what's the current state?
  • Scale resources: add 2 more machines
  • View costs: how much am I spending?
  • Troubleshoot: why is the API returning 500 errors?
  • Access logs: show me API server logs
  • Manage infrastructure: migrate to DigitalOcean
Specialized UI elements:
  • Secure credential prompts - Safe input for cloud provider credentials and API keys
  • Runtime status display - Real-time view of your app’s health and resources
  • Progress indicators - Track long-running operations like deployments
  • Interactive approvals - Confirm destructive actions before execution
Single chat, multiple agents: Monk internally uses multiple specialized agents for different tasks (code analysis, cloud provisioning, monitoring, etc.). You talk to one unified chat. That’s it. No context switching, no juggling agents.

​
Task Board View

The built-in task board shows you everything Monk is doing. What it shows:
  • Active tasks - Deployments, scaling operations, migrations in progress
  • Completed tasks - History of what Monk has done
  • Pending approvals - Actions waiting for your confirmation
  • Task status - Animated progress indicators with real-time updates
Task Board View You see what’s happening at a glance. Long-running deployments, migrations, completed work and outcomes — nothing gets lost.

​
Terminal Integration

Monk opens standard IDE terminals for direct access. Live logs: 
You: Show me API server logs

[Monk opens terminal with live log stream]
Logs appear in a normal terminal window with live updates as events happen. Shell access: 
You: Give me shell access to the database server

[Monk opens terminal with a shell connection via Monk's encrypted P2P network]
Direct shell access to any VM or container in your system. No leaving your IDE. No remembering connection strings.

​
Bug Reporting

Found an issue? Report it straight from the chat window.
  1. Click the bug icon in the chat interface
  2. Describe what went wrong
  3. Monk automatically attaches relevant context (logs, state, recent actions)
  4. Bug report goes to the Monk team
Monk Bug report

​
Interaction with Coding Agents

Monk works directly with coding agents through MCP. If you want to use Monk with GitHub Copilot in VS Code, Claude Code, Gemini, Codex, Cursor, Windsurf, or Antigravity, see:

MCP Getting Started

Connect Monk to your coding agent in minutes.

MCP Support

Full details on Monk’s MCP integration.