mono/packages/kbot/tests/scripted

KBot - Command Line AI Assistant

KBot is a powerful command line tool that leverages AI models to assist with code generation, file summarization, multi-file processing, and more.

Table of Contents

• Installation
• Usage
• Examples
• Architecture
• Configuration
• Development
• License

Installation

1. Ensure you have Node.js installed (version 14+ recommended).
2. Clone the repository:

git clone https://github.com/yourusername/kbot.git
cd kbot

3. Install dependencies:

npm install

4. Build the project (if applicable):

npm run build

Usage

KBot provides a rich set of commands for interacting with AI tools. Here are some common commands:

• Run a task:

kbot run --prompt "Summarize the project" --path ./src

• Fetch available models:

kbot fetch

• Modify configurations:

kbot init

Examples

Below are some usage examples:

1. Summarize Project Files:

kbot run --prompt "Give me a summary of the project files" --path ./my_project --include "*.js"

2. Generate Documentation:

kbot run --prompt "Generate documentation for the codebase" --dst ./docs/README.md

3. Personalized Assistant:

Use your own preferences and profiles stored in ./.kbot/preferences.md for a personalized experience.

Architecture

The following Mermaid diagram illustrates the high-level architecture of KBot:

flowchart TD
Start[Start] --> Config[Load Config]
Config --> Init[Initialize Client]
Init --> Process[Process Request]
Process --> Options[Set Options]
Options --> Execute[Execute Task]
Execute --> End[Return Result]

The flow begins with startup, loads the configuration (preferences and settings), initializes the API client, gathers and processes user requests, sets task-specific options, executes the task (running the completion, tools, or assistant mode), and finally returns the result.

Configuration

KBot uses a configuration file located at ./.kbot/config.json and a preferences file at ./.kbot/preferences.md to customize behavior:

• config.json: Contains API keys and service configurations for OpenAI, OpenRouter, and more.
• preferences.md: Stores personal information to tailor the assistant responses.

Development

• Run tests:

npm test

• Build the project:

npm run build

• Lint the code:

npm run lint

Mermaid Diagrams

Mermaid diagrams are used to visually represent the various components and flows within KBot. Here is another example diagram illustrating the internal processing:

flowchart LR
User[User Input] --> CLI[Command Line Parser]
CLI --> Processor[Task Processor]
Processor --> API[API Client]
API --> Collector[Response Collector]
Collector --> Output[Display Output]

License

This project is licensed under the MIT License.