deepseek-harness-desktop

agent
Security Audit
Warn
Health Warn
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 8 GitHub stars
Code Pass
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
  • Permissions — No dangerous permissions requested

No AI report is available for this listing yet.

SUMMARY

Community-maintained self-contained Tauri desktop distribution for DeepSeek Harness

README.md

DeepSeek Harness

English | 中文

DeepSeek Harness (dsh) is an open-source agent harness developed by DeepSeek AI.

The Tauri desktop installers in this repository are a community-maintained MIT-licensed fork, not an official DeepSeek AI signed distribution. Harness core remains upstream-derived; this repository maintains the desktop carrier, installers, and related support.

It uses an architecture where everything is a plugin, and is powered by Cordis, whose design is described in A Programming Paradigm for Spatiotemporal Composability.

Developer preview

DeepSeek Harness is currently in developer preview and is iterating rapidly. THERE WILL BE COMPATIBILITY-BREAKING CHANGES.

Run

Windows desktop distribution (community build)

This branch provides a self-contained Tauri desktop carrier and Windows x64 MSI. The installer bundles Node.js, pnpm, the production Harness Web/CLI runtime, and the WebView2 offline installer; recipients do not need Node.js, pnpm, Rust, or Git preinstalled.

Download the installer matching your system from GitHub Releases. Windows x64 users can choose DeepSeek-Harness-Desktop-0.1.12-x64-en-US.msi or the Chinese-localized DeepSeek-Harness-Desktop-0.1.12-x64-zh-CN.msi, then verify it with the release's SHA256SUMS.txt.

The bundled Node.js 24 runtime sets the supported Windows floor: x64 releases target Windows 10 / Windows Server 2016 and newer. Windows 11 x64 is locally accepted; Windows 10 x64 is a supported target pending an independent clean-machine pass. Windows ARM64 requires its separately built MSI and remains a preview target until native-machine acceptance is recorded.

On the first launch of each app version and architecture, the desktop host publishes its runtime under %LOCALAPPDATA%\\com.deepseek.harness.desktop\\runtime. Materializing more than thirty thousand physical files can take around two minutes on a PC running Defender and another security product. This is a one-time initialization; later launches reuse the versioned directory and its atomic readiness marker. Failures expose Retry, Copy diagnostics, and Open diagnostics directory actions instead of requiring users to inspect a hidden CMD window.

The carrier preserves Harness profiles, patches, plugins, skills, sessions, and project directories; it does not compile Harness into a non-extensible monolithic executable. See the desktop documentation for source-build instructions and distribution boundaries. The Windows x64 MSI has been exercised locally through cold/warm startup, real Web HTTP, CLI/TUI initialization, bundled pnpm, and owned-process cleanup. macOS, Linux, and ARM64 use native CI build jobs and are not described as stable until they receive equivalent platform testing.

Run from npm

Install Node.js, then run:

npx @deepseek-ai/dsh web

The command starts the Web UI at http://127.0.0.1:3080 by default and opens it in the default browser for a local launch. An SSH launch only prints the host URL because the SSH client or editor owns the local forwarded address. Pass --no-open to run the server without opening a browser. See Web UI guide.

Run from source

To run from a repository checkout:

git clone https://github.com/deepseek-ai/deepseek-harness.git
cd deepseek-harness
pnpm install
pnpm run build
pnpm dsh web

pnpm run build prepares the repository artifacts. pnpm dsh web uses those built artifacts without rebuilding.

Community and support

  • For desktop installation, startup, or update problems, open an issue in this repository and include the Windows version, app version, and diagnostics.
  • For Harness core usage and feature discussions, use the upstream GitHub Discussions.
  • When publishing a plugin, add the dsh-plugin topic so other users can find it.
  • For real-time chat, join the Discord community. Chinese-speaking users can also join the WeCom group below.

Chinese community

Complete the join request form and add the WeCom assistant. After approval, the assistant will send a group invitation. Follow the official WeChat account for project updates.

Show the WeCom assistant, join request, and official WeChat account QR codes
WeCom assistant Join request Official WeChat account
DeepSeek Harness WeCom assistant QR code DeepSeek Harness join request QR code DeepSeek Harness official WeChat account QR code

Contributing

See CONTRIBUTING.md.

Development

Start with the development guide and architecture documentation.

For agents, follow AGENTS.md.

License

MIT

Third-party dependencies and their licenses are disclosed in THIRD_PARTY_NOTICES.md.

Reviews (0)

No results found