AutoCoder — Testing skill for Claude Code

AutoCoder

AutoCoder is an AI-powered desktop coding workspace for end-to-end software delivery. It combines a Tauri desktop shell, a React frontend, and a Rust orchestration backend to coordinate planning, coding, debugging, review, and testing across multiple agents.

This repository is currently an early-stage open-source project. The internal product name `AI Dev Hub` still appears in parts of the UI and codebase.

What This Project Is For

Most AI coding tools are optimized for one agent in one chat. AutoCoder is built around a different idea: software delivery is a workflow, not a single prompt.

This project tries to make that workflow explicit:

• A Director decides which skill should run next
• plan, code, debug, review, and test are treated as separate execution modes
• Claude and Codex are used for different responsibilities
• shared blackboards (PLAN_BLACKBOARD.md, BLACKBOARD.md, BLACKBOARD.json) act as the coordination layer
• project files, docs, history, tool logs, and session state stay visible in the desktop app

In practice, AutoCoder is trying to be a local AI software delivery console rather than a plain chat wrapper.

Core Features

• Multi-stage workflow orchestration with a Director layer
• Dedicated modes for planning, coding, debugging, review, and testing
• Shared blackboard collaboration instead of loose agent-to-agent chat
• Workspace creation, file tree exploration, and project document ingestion
• Session history with restorable Director context
• Tool call logging for Claude and Codex runs
• Inline subtask review loop: Claude implements, Codex reviews, Claude repairs
• Bundled skills system (frontend-dev, fullstack-dev, ui-design-system, simplify, verify)
• Configurable Director backend through OpenAI-compatible or Anthropic-compatible APIs
• Desktop UI built with Tauri, React, TypeScript, and Tailwind CSS

How It Works

1. Director

The Director is the traffic controller. It reads the user's input and