awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
dataelement avatar

dataelement/dsh-desktop

0
View on GitHub↗
dshdesktop.com↗

Dsh Desktop

dsh-desktop is a desktop client that runs a DeepSeek-powered AI agent locally. It starts and stops the bundled agent runtime automatically, so there are no services or ports to manage, and it provides native windows, menus and folder pickers. It checks for new versions at startup, periodically and after resume, and installs an update only after the user consents. User data is kept across upgrades.

The app is built as a plugin-extensible shell. Plugins load per profile and install through a manager that checks the package name, version and registry integrity, and it reports each operation as applied, restart-required, failed, cancelled or overridden. When a plugin blocks startup or rendering, guided recovery traces the failure to the responsible plugin. A safe mode then boots only installer-bundled packages under a separate profile, keeping sessions, models and workspaces intact. Two further differentiators are workbench apps and portable agent presets. Workbench apps are business-focused interfaces that can be browsed, installed from a marketplace or imported locally, and that run beside agent conversations. Custom agent presets export as archives that exclude credentials and sessions. Imports are validated, previewed and conflict-checked, with trust warnings, before anything is installed.

Beyond that, the app connects third-party model providers, with model search and options for image input and reasoning. It generates editable Word, Excel and PowerPoint files from native templates and supplied material, with the needed runtime bundled and custom presentation templates supported. A phone can be paired by QR code to continue a session over the local network or a Cloudflare or Pinggy tunnel, using a temporary password. Websites can also be opened in an isolated sidebar browser with tabs.

Features

  • AI Desktop Clients - Runs a DeepSeek-powered AI agent locally in a desktop app with managed runtime, native windows, folder pickers and automatic updates.
  • Agent Workbench App Platforms - Provides a marketplace and local import for business apps that run beside agent conversations while keeping sessions and project files intact.
  • Portable Agent Presets - Exports custom agent presets as archives without credentials or sessions, and validates, previews and conflict-checks imports with trust warnings before installing.
  • Workbench App Marketplaces - Offers a marketplace to browse and install business-focused workbench apps or import local ones, each paired with an agent session.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
17,546 stars·609 forks·JavaScript·MIT·7 views
  • Plugin Package Managers - Installs plugins through a manager that verifies package name, version and registry integrity and reports applied, restart-required, failed or cancelled outcomes.
  • Bundled Backend Process Supervision - Starts, supervises and stops the bundled agent runtime automatically so users never manage services or ports.
  • Desktop and Native Integration - Starts and stops the bundled agent runtime automatically and provides native menus, windows, folder pickers, update checks and data preserved across upgrades.
  • Plugin-Extensible Application Shells - Loads third-party plugins per profile, verifies installed packages, traces load failures to the responsible plugin, and offers a safe mode that preserves user data.
  • Preset Exporters and Importers - Exports custom agent presets as portable archives holding setup and display details while excluding credentials, sessions and workspace files.
  • User Profile Isolations - Boots a safe mode under a separate profile with only installer-bundled packages, keeping sessions, models and workspaces intact.
  • Safe Mode Profiles - Launches an isolated safe-mode profile with only official core features, keeping the agent and user data and letting users remove third-party plugins.
  • Safe Mode Recoveries - Detects when a third-party plugin blocks startup or rendering and opens guided recovery with a safe mode that keeps sessions, models and workspaces.
  • Agent Preset Packages - Exports custom agent presets as archives without credentials or sessions, and imports them with validation, preview, conflict checks and trust warnings.
  • Remote Model Connectivity - Simplify first-time connection to model providers, search models by name, and enable image input and reasoning options so outside models work reliably.
  • Model Provider Configurations - Sets up third-party model providers with guided first-time setup, model search by name, and image input and reasoning options.
  • Model Provider Settings - Lets users connect third-party model providers, search models by name, and enable image input and reasoning options for compatibility.
  • Automatic Version Updates - Checks for new versions at startup, periodically and after resume, and installs an update only after the user consents.
  • Office Document Generation - Generates editable Word, Excel and PowerPoint files from templates and source material, with the required runtime and dependencies bundled.
  • Task-Specific Workbench Apps - Lets users browse, install or locally import business-focused workbench apps and run them alongside agent conversations.
  • Workbench Apps - Lets users browse, favorite, install or import business-focused apps that run beside agent conversations, keeping sessions and project files when an app is removed.
  • Desktop-to-Phone Companion Pairing - Pairs a phone with the desktop app through QR codes and short-lived tokens over the local network or Cloudflare or Pinggy tunnels.
  • Phone Session Continuation - Lets an ongoing agent session carry on from a paired phone over the local network or an optional internet tunnel.
  • Phone Session Pairing - Pairs a phone by QR code to continue agent sessions over the local network or a Cloudflare or Pinggy tunnel with a temporary password.
  • Sandboxed Browser Runtimes - Opens websites in an isolated sidebar browser with tabs, confining popups, permissions, downloads and backend requests away from the main app.
  • Plugin Failure Recovery - Traces startup or rendering failures to the responsible plugin and offers targeted repairs, asking for consent before destructive profile changes.
  • QR Code Pairing - Pairs a phone with the desktop app by scanning a QR code and continues sessions using a one-time password.
  • Star history

    Star history chart for dataelement/dsh-desktopStar history chart for dataelement/dsh-desktop

    How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

    Frequently asked questions

    What does dataelement/dsh-desktop do?

    dsh-desktop is a desktop client that runs a DeepSeek-powered AI agent locally. It starts and stops the bundled agent runtime automatically, so there are no services or ports to manage, and it provides native windows, menus and folder pickers. It checks for new versions at startup, periodically and after resume, and installs an update only after the user consents. User data is kept across upgrades.

    What are the main features of dataelement/dsh-desktop?

    The main features of dataelement/dsh-desktop are: AI Desktop Clients, Agent Workbench App Platforms, Portable Agent Presets, Workbench App Marketplaces, Plugin Package Managers, Bundled Backend Process Supervision, Desktop and Native Integration, Plugin-Extensible Application Shells.

    Which projects share features with dataelement/dsh-desktop?

    Projects with overlapping indexed features include: fathah/hermes-desktop — Hermes Desktop is a cross-platform desktop application that acts as a graphical companion for installing, configuring,… anywhere-labs/dsh-desktop — dsh-desktop is a native desktop application for the DeepSeek Harness ecosystem. It bundles a local web interface and a… linuxhsj/openclaw-zero-token — OpenClaw Zero Token is a self-hosted gateway and assistant platform that lets people chat with browser-based AI… hardbeat920/monocode — Monocode is a native desktop application, written in Rust, that runs several installed AI coding agent command-line… wklken/k-vim — k-vim is a configuration framework and development environment for the Vim text editor. It provides a set of… ekkolearnai/ekko-studio — Ekko Studio is a self-hosted AI agent workspace that runs several agents together in shared chats and keeps sessions…

    Projects sharing features with Dsh Desktop

    These projects share indexed features with Dsh Desktop. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
    • fathah/hermes-desktopfathah avatar

      fathah/hermes-desktop

      14,548View on GitHub↗

      Hermes Desktop is a cross-platform desktop application that acts as a graphical companion for installing, configuring, chatting with and managing a Hermes AI agent. The agent can run locally on the user's machine or on a remote machine, reached through an API server URL and key or through SSH tunnels. A streaming chat interface shows replies, reasoning, tool progress and live token and cost counts as they arrive, supports slash commands, and lets users search and resume past sessions. Beyond chat, the app works as a management console for the agent. Users can create and switch between isolate

      TypeScriptaiai-agentautonomous-agents
      View on GitHub↗14,548
    • anywhere-labs/dsh-desktopanywhere-labs avatar

      anywhere-labs/dsh-desktop

      35,279View on GitHub↗

      dsh-desktop is a native desktop application for the DeepSeek Harness ecosystem. It bundles a local web interface and a background service into a single app that starts, supervises, and stops the service automatically, with a system tray and no command-line setup. It replaces workflows that previously required the command line with a graphical interface for launching and managing DSH plugins. The app is built entirely from plugins. The desktop shell, windows, tray, terminal, and updater are all plugins that combine with third-party plugins through one shared mechanism, so the shell itself can

      TypeScriptcordiscordis-plugindeepseek
      View on GitHub↗35,279
    • linuxhsj/openclaw-zero-tokenlinuxhsj avatar

      linuxhsj/openclaw-zero-token

      5,213View on GitHub↗

      OpenClaw Zero Token is a self-hosted gateway and assistant platform that lets people chat with browser-based AI services such as ChatGPT, Claude, Gemini, DeepSeek, Kimi and Grok without paying for API tokens. It captures a login from a debuggable Chrome browser through the Chrome DevTools Protocol, then reuses that session's cookies and headers for later requests. A single OpenAI-compatible chat completions endpoint, secured by a bearer token, routes each request to whichever web model is configured, so outside programs can call those models like a standard API. Because web chat models have n

      TypeScript
      View on GitHub↗5,213
    • hardbeat920/monocodehardbeat920 avatar

      hardbeat920/monocode

      3,737View on GitHub↗

      Monocode is a native desktop application, written in Rust, that runs several installed AI coding agent command-line tools from one interface. It reuses each tool's existing sign-in and the user's own subscriptions rather than selling tokens. Sessions live in tabs and split panes grouped by project, and each parallel task can run in its own Git worktree, so several agents can work on the same project without overwriting each other's files. Beyond running a single agent, it supports splitting work across up to four worker agents, handing a session to another model with a recap, and continuing p

      Rust
      View on GitHub↗3,737
    Compare all 30 related projects→