1.5 KiB
1.5 KiB
Product Guide
Initial Concept
A robust, modular, and high-performance AI coding assistant for the terminal, built in Rust. It aims to streamline software development workflows by integrating directly with the codebase and version control systems.
Target Audience
- Software Developers: The primary users who need assistance with coding tasks, refactoring, and understanding complex codebases.
- DevOps Engineers: For handling infrastructure-as-code and script automation via natural language.
Core Value Proposition
- Performance & Safety: Leverages Rust's memory safety and speed for a responsive CLI experience.
- Model Agnostic: Supports multiple LLM backends (Anthropic, OpenAI, Ollama) giving users flexibility.
- Deep Integration: Deeply integrates with the local development environment (filesystem, git, shell).
- Extensibility: A plugin architecture to extend functionality with custom commands and agents.
Key Features
- Natural Language Interface: Interact with the agent using plain English to perform complex tasks.
- Codebase Awareness: The agent can read and analyze the project structure and file contents to provide context-aware assistance.
- Multi-LLM Support: Configurable providers including Anthropic, OpenAI, and local models via Ollama.
- Rich TUI: A text-based user interface for a better user experience in the terminal.
- Tool Ecosystem: A suite of built-in tools for file manipulation, shell execution, and more.