DotHarness/dotcraft

The agent harness for .NET: embeddable, extensible, remotely controllable and ready to use.

What is DotCraft?

DotCraft is an open‑source AI‑agent runtime you can run on your own computer or server. It ships as a desktop application, a command‑line tool, mobile app, and a set of SDKs (TypeScript and .NET) so developers can embed the agent in their own products. The agent can plan work, break it into sub‑tasks, call tools, and even control other desktop applications ("Computer Use").

Key Features

Feature What it does
Cross‑device workspace Desktop, CLI, IDE, phone and chat bots share a single workspace, so a task started in one place can be continued elsewhere.
Sub‑agents The main agent can spawn helper agents that work on parts of a task in parallel.
In‑app browser & Computer Use The agent can open a web page or interact with native Windows/macOS apps you allow it to control.
Agent Builder Turn a natural‑language description into a reusable agent profile.
Automations, Goals, Dreams, Dynamic Workflows Define long‑running behaviours that keep work moving without your constant attention.
Oratorio board Integrated project board that aggregates local tasks, GitHub/GitLab issues and PRs, letting you hand cards to the agent.
Chat‑bot channels Connect the agent to Feishu/Lark, QQ, Telegram, WeCom, WeChat so teammates can ask it questions from group chats.
Satellite Run the agent on a second computer (e.g., a colleague’s PC) while keeping control and approval dialogs on the host machine.
Harness library .NET library (DotCraft.Harness) that exposes the full runtime (sessions, tools, approvals, model providers) for embedding in any .NET app.
SDKs TypeScript (@dotcraft/sdk) and .NET (DotCraft.Sdk) packages let you programmatically start threads, run turns, and stream results.
Avatar package React component (@dotcraft/avatar) that renders a customizable robot avatar reflecting the agent’s state.

How to Get Started

  1. Desktop – Download the installer for Windows, macOS or Linux from the releases page, open a folder as a workspace, and connect an LLM provider (any OpenAI‑compatible or Anthropic‑compatible API, or sign in with a ChatGPT subscription).
  2. CLI – Run a one‑liner install script (curl … | bash on macOS/Linux or PowerShell irm … | iex on Windows) and use dotcraft exec "<task>".
  3. Mobile – Install the Android app, scan the QR code shown in Desktop’s Settings → Connections → Phones, and monitor or approve tasks from your phone.
  4. Embedding – Add the NuGet package DotCraft.Harness to a .NET project or the npm package @dotcraft/sdk to a JavaScript/TypeScript app, then follow the quick‑start code snippets in the README.

Who Might Use It?

  • Developers who want an on‑premise AI assistant to automate code reviews, refactorings, testing, or any repetitive dev‑ops work.
  • Product teams that need a shared AI‑agent across desktop, mobile, and chat platforms for collaborative task management.
  • Companies that require control over model providers and API keys, keeping secrets on a single host while multiple users interact with the agent.
  • Tool builders who want to extend the runtime with custom .NET plugins, React desktop plugins, or their own SDK‑based integrations.

Where to Learn More

  • Documentation site – https://www.dotcraft.net/
  • Quick‑start guides – Desktop, CLI, Mobile, SDKs, Harness, Channels, Satellite, Oratorio.
  • Contribution guide – CONTRIBUTING.md in the repo.
  • License – Apache 2.0 (permissive, commercial‑friendly).

All information is taken directly from the repository’s README; no external assumptions have been added.

Related

  • Project
    Ebony-Vinyl/EAC-Desktop

    An open-source desktop workbench based on DeepSeek Harness that integrates AI agent capabilities into a native desktop environment with bundled runtimes and a plugin ecosystem.

  • Project
    dotnet/skills

    A repository of standards‑compliant, AI‑driven plugins (called *skills*) that let large‑language‑model coding assistants perform concrete .NET development tasks—building, debugging, upgrading, testing, UI work, and even ML integration—across Copilot, Claude, Cursor, and OpenAI Codex. Includes a public metrics dashboard and ready‑made agentic workflows.

  • Project
    clawdotnet/openclaw.net

    OpenClaw.NET is a .NET‑native, self‑hosted AI‑agent runtime with a desktop Avalonia UI (AgentQi Companion). It runs a gateway that talks to any OpenAI‑compatible model, offers streaming, tool execution, session memory, and many integrations (REST, WebSocket, MCP, 80+ tools, channel adapters). The gateway can be compiled with NativeAOT, making it suitable for edge or on‑prem deployments. Users can start via a ready‑made desktop bundle or from source, configure the assistant through chat commands, and extend functionality with .NET plugins or reusable MetaSkill DAG workflows.

  • Project
    SciSharp/BotSharp

    An open-source .NET framework for building AI bot platforms that allows enterprise developers to integrate LLMs and multi-agent systems into business applications.

  • Project
    feder-cr/invisible_dots

    invisible_dots is an open-source, self-hosted AI agent platform where each agent (a "Dot") lives in its own hardware-accelerated virtual machine with a patched Firefox browser that evades anti-bot detection. It allows long-running tasks, scheduled automations, and integration via MCP servers, CLI, web UI, and messaging channels.