# dataelement/dsh-desktop

DSHDesktop：DeepSeek Harness Desktop / DeepSeek Harness 桌面版

Repository: https://github.com/dataelement/dsh-desktop
Canonical: https://ross.abutalabs.com/products/dataelement-dsh-desktop
Homepage: https://dshdesktop.com
Language: TypeScript
License: MIT
License Family: permissive
Topics: deepseek, desktop, dsh-plugins, dshdesktop, deepseek-harness, deepseek-harness-desktop, harness, agent, apps
Last push: 2026-08-26T15:33:38+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 1
- inputs: {"age_days": 20, "days_push": 7, "days_rel": 7, "gap_med": 0, "n_releases_24m": 20}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2691, forks 161 (observed 2026-08-28T04:07:09.460607+00:00)

## What it is
DSH Desktop is a local-first, cross-platform desktop application that packages the DeepSeek Harness agent runtime as an installable app for macOS and Windows. It automatically starts and manages the local Harness instance, handles model provider configuration, and keeps profiles, plugins, workspaces, and sessions outside the install directory.

## Use cases
- run DeepSeek Harness as a desktop app without Node or CLI
- chat with DeepSeek and third-party models like OpenAI, Anthropic, and Gemini locally
- manage agent presets and project workspaces with a native directory picker
- continue agent sessions from a phone
- install an AI agent runtime with one click and no manual model ID configuration
- keep API keys in a local credential store while using multiple model providers

## When to choose
- you want DeepSeek Harness as a ready-to-use desktop application
- you need a no-setup agent runtime supporting mainstream model providers
- you want sessions, plugins, and settings preserved across app upgrades

## When to avoid
- you need Linux support, which is not offered
- you prefer a headless CLI or server deployment of the agent runtime
- you need a fully stable release rather than an early preview

## Facets
- artifact type: application
- maturity: experimental
- function: agent-framework, chat-interface, gui, llm-inference, developer-tools
- domain: large-language-models, chatbots, desktop-applications, developer-tools, cross-platform
- platform: windows, cross-platform
- tags: deepseek, deepseek-harness, desktop-app, agent-runtime, model-providers, local-first, electron, portable-presets, phone-control, ai-agents, macos, desktop

## Member repositories
- dataelement/dsh-desktop (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:09.460607+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:16:33.351676+00:00, confidence not recorded.
  - readme: https://github.com/dataelement/dsh-desktop (fetched 2026-08-28T04:07:09.460607+00:00, sha 73908723c29a)
  - homepage: https://dshdesktop.com (fetched 2026-08-29T09:59:51.462727+00:00, sha 0ee0995eca91)
- Data as of 2026-08-30T08:39:29.467469+00:00.