Installing Kilocode Beginner
Get Kilocode running in VS Code in minutes. This lesson covers installation from the marketplace, API key setup, running your first task, understanding the interface, and the permission system.
Installing from VS Code Marketplace
Method 1: From VS Code
-
Open VS Code
Launch Visual Studio Code on your machine.
-
Open the Extensions panel
Press
Ctrl+Shift+X(Windows/Linux) orCmd+Shift+X(macOS). -
Search for "Kilocode"
Type
Kilocodein the search bar. Look for the official Kilocode extension. -
Click Install
Click the blue Install button. The extension activates automatically.
Method 2: From the Command Line
# Install Kilocode via the VS Code CLI code --install-extension kilocode.kilocode
API Key Setup
Kilocode supports the same providers as Cline. Set up your preferred AI provider:
| Provider | Setup Steps |
|---|---|
| Anthropic | Get an API key from console.anthropic.com. Paste into Kilocode settings. |
| OpenAI | Get an API key from platform.openai.com. Select OpenAI provider in settings. |
| OpenRouter | Get an API key from openrouter.ai. Access all models with one key. |
| Get an API key from Google AI Studio. Select Gemini provider in settings. | |
| Local (Ollama) | Install Ollama, pull a model, select Ollama in Kilocode settings. Free and private. |
Your First Task
-
Open the Kilocode panel
Click the Kilocode icon in the VS Code sidebar or use the command palette (
Ctrl+Shift+Pand type "Kilocode"). -
Open a project
Make sure you have a project folder open. Kilocode works within your current workspace.
-
Type a task
Start with something simple: "Explain the structure of this project and its main dependencies."
-
Review and approve
Kilocode will ask permission before reading files. Approve each action to let it proceed.
You: Explain the structure of this project and list the main dependencies. Kilocode: I'll analyze the project structure... [Read package.json] Approve | Reject [Read tsconfig.json] Approve | Reject [List src/ directory] Approve | Reject This is a TypeScript Express API with the following structure... // Cost: $0.003 | Model: Claude Haiku (auto-selected)
Understanding the Interface
Kilocode's interface builds on Cline's layout with several enhancements:
| Area | Description | Enhanced in Kilocode |
|---|---|---|
| Chat input | Type tasks, drag-and-drop files and images | Smart file suggestions as you type |
| Diff view | Review proposed code changes | Enhanced syntax highlighting, inline annotations |
| Cost display | Running cost of current task | Per-model breakdown, budget progress bar |
| Model selector | Choose the AI model | Auto-routing option, model recommendations |
| Task history | View and resume previous tasks | Search, filter, and cost summary per task |
| Settings | Configure providers and behavior | Cost dashboard, routing rules, budget alerts |
Permission System
Like Cline, Kilocode asks for approval before taking actions. The permission categories are:
- Read files - Reading workspace files for context
- Write/Edit files - Creating or modifying files
- Run commands - Executing terminal commands
- Use browser - Launching and controlling a browser
- Use MCP tools - Calling external MCP server tools
Each can be set to "Ask each time" or "Auto-approve." Kilocode adds the ability to set auto-approve limits per action type and per model.
Quick Start Guide
Here's a 5-minute getting-started checklist:
- Install Kilocode from the VS Code marketplace
- Open Kilocode settings and add your API key
- Open a project folder in VS Code
- Click the Kilocode icon in the sidebar
- Type: "What does this project do?"
- Approve the file read requests
- Try a simple edit: "Add a comment to the main entry file explaining what it does"
- Review the diff and approve if it looks good
Kilocode Installed?
With Kilocode set up, explore its enhanced features in the next lesson - including smart diff views, model routing, and the cost dashboard.
Next: Features →Ready to Go Deeper?
Live instructor-led courses from our partners. Affiliate disclosure.
AI & ML Courses - 30% Off
Live instructor-led AI, machine learning, data science, and cloud courses for working professionals. Use code Limited30 at checkout.
EdurekaDataCamp - AI & Data Science
Hands-on Python, machine learning, and AI courses with interactive exercises and real projects.
DataCampedX - Top AI Courses
University-level AI courses from MIT, Harvard, Stanford. Earn certificates that employers recognize.
edX