GoDiao/dreamcoder

完全开源claude desktop的桌面编程工作台| https://godiao.github.io/dreamcoder/

DreamCoder – Desktop GUI for Claude Code

What it is – DreamCoder is an open‑source, native‑desktop application that wraps the Claude Code AI‑coding model (Anthropic’s code‑focused LLM) in a graphical interface. It lets you chat with the model, run code, manage sessions, and invoke tools (file system, terminal, screenshots) without leaving a desktop window.

Core capabilities (as described in the README)

Feature What you get
Native desktop experience Multi‑tab conversation view, visual history, side‑panel navigation, built‑in PTY (PowerShell/Bash/Zsh) displayed via xterm.js.
Claude Code integration Re‑uses Claude Code’s core logic (or a compatible runtime) and adds two Computer Use modes: a screenshot‑based UI and a cheaper UIA Tree text‑based mode.
Transparent tool calls File reads/writes and terminal commands executed by the model are shown in the UI so you can audit what the AI does.
Model‑provider switching One‑click selection among many LLM providers – Anthropic (Claude), OpenAI, DeepSeek, Moonshot/Kimi, MiniMax, Azure OpenAI, Google Vertex, AWS Bedrock, etc.
MCP (Model Context Protocol) support Full MCP integration with a graphical configuration UI, allowing extensions of the model’s context and tool set without hand‑editing JSON.
Privacy‑first API keys and conversation data are stored locally; no cloud‑hosted backend is required.

Technical stack (from the README)

  • Desktop shell – Tauri 2 (Rust) – gives a small, native binary.
  • Frontend UI – React 18 + Vite + Tailwind CSS 4, state managed with Zustand.
  • Backend runtime – Bun (Node‑compatible) for side‑car processes.
  • Terminal – portable-pty (Rust) + xterm.js.
  • Communication – WebSocket, Model Context Protocol (MCP), Language Server Protocol (LSP).

Current platform support

OS Status Install package
Windows x64 Actively tested by maintainers NSIS .exe and MSI .msi provided in each release
macOS arm64 Not regularly verified (scripts present) No pre‑built package yet
Linux x64 Not regularly verified No pre‑built package yet

The code contains conditional branches for macOS/Linux, but the maintainer focuses on Windows. Community help is invited to improve those platforms.

Getting started (quick‑start from the README)

# 1. Clone the repo
git clone https://github.com/GoDiao/dreamcoder.git && cd dreamcoder

# 2. Install root dependencies (Bun)
bun install

# 3. Install desktop‑app dependencies
cd desktop && bun install

# 4. Build side‑car binaries (Anthropic SDK, AWS SDK, etc.)
bun run build:sidecars

# 5. Run the app in development mode
bun run tauri dev

Prerequisites: Bun ≥ 1.0, Rust toolchain, Node ≥ 18. Linux users must also install WebKitGTK, libappindicator, librsvg, etc. (see Tauri docs).

Roadmap highlights

  • Phase 1‑2 – Completed: GUI, multi‑model support, Computer Use, MCP, skill extensions.
  • Phase 2.5 – Performance tweaks (bundle splitting, throttling, LRU terminal cache).
  • Phase 3 – Remote H5 access (phone/browser client).
  • Phase 4 – IM adapters (Feishu, DingTalk, Telegram, WeChat).
  • Phase 5 – Automated releases & auto‑update.

Community & contribution

  • Good‑first‑issue and help‑wanted labels are actively maintained; each has a mentor.
  • Contribution guide: docs/CONTRIBUTING_zh.md.
  • License: MIT (2024‑2026).

Bottom line – DreamCoder is a genuine, actively‑maintained open‑source project that brings Claude Code’s LLM capabilities to a user‑friendly desktop environment, with extensible model/provider support and a focus on privacy.

Related

  • Project
    Vasco0x4/AIDA

    AIDA is an open‑source autonomous pentesting platform that lets any LLM (Claude, Gemini, OpenAI, etc.) drive a Docker‑based security‑testing environment. The agent can run reconnaissance, scanning, custom Python exploits, and raw HTTP manipulations, logging every command and its output in a persistent notebook. Findings are auto‑scored with CVSS 4.0 and can be exported as PDF reports. The project includes a web dashboard, JWT authentication, notification hooks, and supports swapping the built‑in `aida-pentest` container for the larger Exegol image. Real CVEs discovered by AIDA are listed, showing practical impact. Licensed under AGPL‑v3.

  • Project
    TaoLiveAIGC/TaoMate

    TaoMate is a research‑grade, real‑time audio‑video digital‑human generation system. The repo provides inference code, multi‑GPU launchers, and a four‑GPU interactive browser demo. It stitches together a 22 B LTX‑2.3 transformer, a Gemma‑3 text encoder, and optional Gemma‑4 dialogue model, requiring high‑memory NVIDIA GPUs and separate model checkpoints from Hugging Face.

  • Project
    allgpt-co/QuickVoice

    QuickVoice is an open‑source, self‑hostable platform for building AI voice agents (inbound/outbound calls) with full control over the stack—marketing site, console, Express API, Python LiveKit worker, telephony (Twilio/Telnyx), RAG (Pinecone), billing (Stripe), and privacy features. It runs locally via a single `task up:dev` command and is designed for teams that need inspectable, extensible phone‑automation infrastructure.

  • Project
    lllllllama/RigorPilot-Skills

    RigorPilot Skills is an open‑source toolkit that lets an AI‑assistant or CLI automatically run the commands described in a research paper’s README, capture the exact output, and insert a concise evidence card back into a side‑by‑side annotated README. It preserves the original file byte‑for‑byte, stores logs, command traces, and optional model‑runner checkpoints in a structured `repro_outputs/` bundle, and provides a suite of reusable “skills” (reproduce, analyze, safe‑debug, explore, etc.) for reproducible AI‑research workflows.