Tiny Trae AI Coding Assistant

An AI coding assistant with tool integration that helps with software development tasks including writing code, debugging, refactoring, and explaining code.

Features

• 🤖 AI-powered coding assistance
• 🔧 Integrated development tools (file operations, search, shell commands)
• 📁 Project-aware file management
• 🔍 Advanced code search capabilities
• 🎨 Modern TUI interface with rich visual feedback (default)
• 📝 Classic CLI interface (optional with --cli flag)
• ⚙️ Configurable system prompts
• 🎯 Support for multiple AI providers (DeepSeek, Anthropic Claude, etc.)

Installation

Quick Install (Recommended)

Unix/Linux/macOS

curl -sSf https://raw.githubusercontent.com/your-username/tiny-trae/main/install.sh | bash

Windows

powershell -c "iwr https://raw.githubusercontent.com/your-username/tiny-trae/main/install.bat -OutFile install.bat && install.bat"

Manual Installation

Prerequisites

• Rust (latest stable)
• Git

Steps

1. Clone the repository:

   git clone https://github.com/your-username/tiny-trae.git
   cd tiny-trae
   

2. Install globally:

   make install
   

   Or using cargo directly:

   cargo install --path .
   

3. Set up configuration: The installer will create a config file at ~/.config/tiny-trae/config.toml. Update your API key:

   [anthropic]
   api_key = "your-api-key-here"
   

Alternative Installation Methods

From crates.io (Recommended)

cargo install tiny-trae-ai

From Source (Development)

git clone https://github.com/your-username/tiny-trae.git
cd tiny-trae
cargo install --path .

Development Installation

git clone https://github.com/your-username/tiny-trae.git
cd tiny-trae
make install-dev

Note: tiny-trae is not yet published to crates.io, so installation must be done from source.

Configuration

Tiny Trae looks for configuration in the following order:

1. Custom config file (if specified with -c flag)
2. Global config: ~/.config/tiny-trae/config.toml
3. Local config: ./config.toml

Configuration Options

[anthropic]
api_key = "your-api-key"
base_url = "https://api.deepseek.com"  # or https://api.anthropic.com

[model]
version = "deepseek-coder"  # or claude-3-5-sonnet-20241022
max_tokens = 4096
temperature = 0.7

[workspace]
root_path = "."
ignore_patterns = [".git", "target", "node_modules"]

[system]
prompt = "Your custom system prompt here"

Usage

Basic Usage (Modern TUI Interface)

tiny-trae

The modern TUI interface provides:

• 🎨 Beautiful visual layout with title bar, message area, status bar, and input area
• 💬 Rich message display with user/assistant differentiation
• 🔧 Real-time tool execution feedback with visual indicators
• ⚡ Instant loading states: User messages appear immediately, followed by AI thinking animation
• 📊 System status and help dialogs
• ⌨️ Intuitive keyboard navigation (↑↓ to scroll, Enter to send, Ctrl+C to exit)
• 🌍 Full Unicode support including Chinese, Japanese, Korean, and emoji input

Classic CLI Interface

tiny-trae --cli

With Custom Config

tiny-trae --config /path/to/config.toml

Verbose Mode

tiny-trae --verbose

Available Commands

In the TUI interface:

• Type help - Show available commands and keyboard shortcuts
• Type history - Show conversation history
• Type clear - Clear conversation history
• Type status - Show system status
• Type exit - Exit the application
• Press Esc - Close dialogs and popups
• Use ↑↓ - Scroll through messages
• Use Page Up/Down - Fast scroll

In the CLI interface:

• help - Show available commands and tools
• history - Show conversation history
• clear - Clear conversation history
• status - Show system status
• tools - List available tools
• exit - Exit the application

Available Tools

• read_file - Read file contents
• list_files - List files in directories
• edit_file - Edit file contents
• search_text - Search for text patterns
• ripgrep - Fast text searching with ripgrep
• bash - Execute shell commands

Examples

Code Analysis

> Can you analyze the main.rs file and suggest improvements?

File Operations

> Read the README.md file and summarize its contents

Code Generation

> Create a new Rust function to parse JSON configuration files

Debugging

> Help me debug this error in my code [paste error]

Uninstallation

cargo uninstall tiny-trae-ai

Configuration files in ~/.config/tiny-trae/ are preserved.

Development

Building from Source

git clone https://github.com/your-username/tiny-trae.git
cd tiny-trae
cargo build --release

Running Tests

cargo test

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

MIT License - see LICENSE file for details.

Support

• 📖 Documentation: GitHub Wiki
• 🐛 Bug Reports: GitHub Issues
• 💬 Discussions: GitHub Discussions

Changelog

See CHANGELOG.md for version history.