CodeBuddy: Your AI Coding Assistant

CodeBuddy is a Visual Studio Code extension that enhances developer productivity through AI-powered code assistance. It provides intelligent code review, refactoring suggestions, optimization tips, and interactive chat capabilities powered by multiple AI models including Gemini, Groq, Anthropic, and Deepseek.

The extension leverages advanced language models to help developers write better code, understand complex codebases, and automate routine tasks. Key features include code commenting, refactoring suggestions, performance optimization, error fixing, code explanation, commit message generation, and interactive code chat. CodeBuddy integrates seamlessly with VS Code's interface and supports TypeScript/JavaScript development workflows with extensibility for other languages.

Install in Vscode Market Place

https://marketplace.visualstudio.com/items?itemName=fiatinnovations.ola-code-buddy

Architecture

flowchart TD

    %% VS Code Extension Frontend
    subgraph "VS Code Extension Frontend"
        CE["Editor Interface"]:::frontend
        CL["Commands Layer"]:::frontend
        WV1["Webview (VS Code)"]:::frontend
        WV2["Webview (React UI)"]:::frontend
    end

    %% Core Application Layer
    subgraph "Core Application"
        AA["AI Agents"]:::core
        MS["Memory System"]:::core
        subgraph "Application Services"
            BL["Business Logic"]:::core
            AI["Application Interfaces"]:::core
        end
        subgraph "Infrastructure Layer"
            HTTP["HTTP Services"]:::infra
            LOG["Logging"]:::infra
            REP["Repository"]:::infra
            LS["Local Storage"]:::infra
        end
    end

    %% AI Providers
    subgraph "AI Providers"
        LLM["Language Model Integrations"]:::provider
        ESP["External Service Providers"]:::provider
    end

    %% Storage Layer
    subgraph "Storage Layer"
        DB["Database (SQLite)"]:::storage
        FS["File System"]:::storage
        VD["Vector Database"]:::storage
    end

    %% Connections between VS Code Extension Frontend
    CE -->|"UserInput"| CL
    CL -->|"ProcessRequest"| AA
    CL -->|"UIUpdate"| WV1

    %% Connections within Core Application
    AA -->|"ContextManagement"| MS
    AA -->|"Orchestration"| BL
    AA -->|"ContractCall"| AI

    %% Connections to Infrastructure
    AA -->|"APIRequest"| HTTP
    HTTP -->|"APICall"| LLM
    HTTP -->|"APICall"| ESP
    HTTP -->|"Feedback"| WV1

    %% Connections from Application Services to Storage
    BL -->|"DataAccess"| DB
    BL -->|"FileAccess"| FS
    BL -->|"EmbedData"| VD
    AI -->|"RepositoryAccess"| REP

    %% Memory System stores context to Storage
    MS -->|"StoreContext"| DB

    %% Styling Classes
    classDef frontend fill:#cce5ff,stroke:#2a6592,stroke-width:2px;
    classDef core fill:#d4edda,stroke:#155724,stroke-width:2px;
    classDef infra fill:#f8d7da,stroke:#a71d2a,stroke-width:2px;
    classDef provider fill:#fff3cd,stroke:#856404,stroke-width:2px;
    classDef storage fill:#d1ecf1,stroke:#0c5460,stroke-width:2px;

    %% Click Events
    click CE "https://github.com/olasunkanmi-se/codebuddy/blob/main/src/extension.ts"
    click CL "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/commands"
    click WV1 "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/webview"
    click WV2 "https://github.com/olasunkanmi-se/codebuddy/tree/main/webviewUi"
    click AA "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/agents"
    click MS "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/memory"
    click BL "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/services"
    click AI "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/application"
    click HTTP "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/infrastructure/http"
    click LOG "https://github.com/olasunkanmi-se/codebuddy/blob/main/src/infrastructure/logger/logger.ts"
    click REP "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/infrastructure/repository"
    click LS "https://github.com/olasunkanmi-se/codebuddy/blob/main/src/infrastructure/storage/local-storage.ts"
    click LLM "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/llms"
    click ESP "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/providers"
    click DB "https://github.com/olasunkanmi-se/codebuddy/tree/main/src/infrastructure/repository"
    click FS "https://github.com/olasunkanmi-se/codebuddy/blob/main/src/services/file-system.ts"

Loading

Database

• SQLite database for code pattern storage
• Vector embeddings for semantic code search

File System

• Workspace management for multi-root projects
• TypeScript configuration detection
• File watching and indexing services

Roadmap

• Codebase Understanding: Comprehensive local codebase indexing with Retrieval-Augmented Generation (RAG) capabilities.
• Code search: Search code across the entire codebase
• Rewrite the Webview UI with React
• Incorporate AI agents for seamless interaction with external resources and APIs.
• Context Pinning: Allow users to pin specific directories, files, and code elements (functions, classes, etc.) as persistent context for AI models.
• Automated Documentation Generation: Generate comprehensive and up-to-date codebase documentation.
• Intelligent Orchestration: Coordinate AI agents' activities with proper workflow.
• Access to real-time data.
• Support for local LLMs such as Ollama
• Support for Deepseek model

Repository Structure

.
├── src/                         # Source code directory
│   ├── agents/                  # AI agent implementations and orchestration
│   ├── application/             # Core application constants and interfaces
│   ├── commands/                # Command implementations for VS Code extension
│   ├── infrastructure/          # Core infrastructure components (HTTP, logging, storage)
│   ├── llms/                    # Language model integrations (Anthropic, Gemini, Groq)
│   ├── memory/                  # Memory management for AI context
│   ├── providers/               # Provider implementations for different services
│   ├── services/                # Business logic and core services
│   └── webview/                 # VS Code webview implementation
├── webviewUi/                   # React-based UI for the extension
└── package.json                 # Project configuration and dependencies

Contribution

1. Fork the repository. Make a branch from Development.
2. npm install the packages
3. Click Run, an option on your screen's top bar, then click Start Debugging. Run -> Start Debugging
4. A new instance of VS Code will be opened up. This is like your test instance; all changes made on the Codebuddy repo can be seen on this instance.
5. The main entry into the application is the extension.ts. For the Webview, the main entry is the webview.tsx
6. Raise a PR
7. For more info, check the contributing.md page

Usage Instructions

Installation

1. Ensure you have Visual Studio Code version 1.78.0 or higher installed.
2. Install the CodeBuddy extension from the Visual Studio Code Marketplace.

Configuration

1. Open VS Code settings (File > Preferences > Settings).
2. Search for "CodeBuddy" in the settings search bar.
3. Configure the following settings:
   • Select the Generative AI model (Gemini, Groq, Anthropic, or XGrok)
   • Enter the API key for your chosen model
   • Choose your preferred font family and chat view theme

Troubleshooting

1. API Key Issues:

   • Problem: "Failed to generate content" error message.
   • Solution: Double-check your API key in the CodeBuddy settings.

2. Model Selection:

   • Problem: Features not working as expected.
   • Solution: Ensure you've selected the correct AI model in the settings.

3. Performance Issues:

   • Problem: Slow response times from CodeBuddy.
   • Solution: Check your internet connection and consider switching to a faster AI model.