Kilo Code

Install Kilo Code Extension

1. 1. Please open VS Code
2. 2. Go to the Extensions view (Or Click Ctrl + Shift + X)
3. 3. Search for "Kilo Code"
4. 4. Click Install on the Kilo Code extension

Open Kilo Code Settings

1. 1. Click on the Kilo Code icon in the VS Code sidebar
2. 2. When you first open Kilo Code, you will see the following prompt. Please select “Use your own API key” or click the settings icon in the top-right corner.

Configure API Provider

1. 1. In the API Provider dropdown, select OpenAI Compatible
2. 2. Enter the following details:
• Base URL: https://inference.canopywave.io/v1
• API Key: Your CanopyWave API key
• Model: Choose a model (e.g.moonshotai/kimi-k2-thinking, deepseek/deepseek-chat-v3.2). See provider-specific routing for more options.

After filling in the URL and key, scroll to the bottom and click the “Let's go” button to complete the configuration.

Test the Integration

1. 1. Enter a message, such as “Please create a tech-inspired webpage for me and provide me with the HTML file directly.”
2. 2. Kilo Code should respond and offer to create the file

All requests will be forwarded through the CanopyWave going forward.

Advanced AI Workflow

1. Multi-Mode Intelligent Switching: Supports specialized modes such as Architect (planning mode), Coder (coding mode), Debugger (debugging mode), and Orchestrator (coordination mode), with seamless switching or custom mode creation for efficient planning and execution of complex tasks.
2. Deep Contextual Understanding: Automatically performs in-depth scanning of the entire codebase to understand project architecture, dependencies, and semantic context, enabling handling of large-scale projects without manual configuration.
3. Automatic Error Recovery and Iteration: Real-time error detection, execution of test suites, and automatic observe-fix loops upon failure, supporting autonomous advancement of multi-step tasks.

Browser Capabilities

1. Browser Automation Interaction: Automatically controls the browser with user approval for operations like clicks and inputs, integrating terminal command execution with browser tasks.
2. Visual and Runtime Analysis: Supports capturing webpage screenshots for visual feedback analysis and real-time reading of browser console logs to automatically detect and fix frontend runtime errors.

Extensibility & Integration

1. MCP Protocol Support (Model Context Protocol): Built-in MCP server marketplace for easy discovery and connection to external tools (e.g., search, GitHub Issues, database queries, document retrieval), significantly expanding AI capabilities.
2. Custom Modes and Rules: Supports creation of unlimited custom modes and definition of project-specific rules via configuration files, ensuring AI adherence to coding styles, tech stacks, or architectural specifications.
3. Context Persistence and Optimization: Memory Bank mechanism for cross-session persistence of project context, intelligently managing long conversation records to optimize token usage while retaining critical development memory.