Media the agent can actually work with: media_probe (ffprobe's reading of any image, video or audio file, translated into a report and a browser-playability verdict with the exact ffmpeg command that would fix it), media_run (a real ffmpeg or ffprobe command as an ARGV array - no shell, a deadline, an output cap, and no overwrite unless asked), and media_frames (frames out of a video at named timestamps, sampled evenly, or tiled into one contact sheet). Ships the pin for a static ffmpeg/ffprobe copy per platform (SHA-256 verified, installed into $DSH_HOME/dsh-media/bin on first need, used only when the machine has none of its own), two bundled skills (ffmpeg-cli, ffprobe-cli), and the host routes dsh-video's tab reads: a Range-capable file stream, the same probe summary as JSON, and a background remux/transcode that turns a container a browser cannot open into a cached MP4.
DSH web plugin: a live, bidirectional mind map built on the mind-elixir-core kernel. The agent and the user edit the same tree in real time through a session projection and a Connection RPC channel.
Agent collaboration · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:SSShooter/dsh-mindmap-live
Windows 极简模式 / Windows Minimal Mode: persistent bash + str_replace_editor dsh plugin — registers agent presets (minimal-msys2, minimal-extended), provides a working persistent-bash backend on Windows, exposes a GUI-editable bash path in the plugin-settings section (credentials domain), adds a proper tool-call row for str_replace_editor, and a minimal→standard progressive preset.
Agent collaboration · Interface extensions
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:AythyaCrispus/dsh-minimal-msys2
Cross-session memory with governance, self-evolution and an approval-gated write path for DeepSeek Harness. Heat-ranked recall, environment conventions, forgetting lifecycle, supersession chains, poisoning defense, auto-distillation, and an open measurement ABI.
Memory and knowledge · Agent collaboration · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Rottenwooood/dsh-mnemos
Modal: the pack's shared dialog surface for the vncode Web GUI. A body-level overlay any client plugin reaches through ctx.get('modals') - open/alert/confirm/prompt, with async submit work that keeps the dialog open and shows the failure inline - plus a RICH dialog (spec.content plus size: 'lg'), a render function handed { close } that owns the whole body for a dialog that is not a question, whose mask click deliberately does not close it. Alpha.
DeepSeek Harness plugin: shows model usage and account balance in the session header — OpenCode Go plan usage (rolling / weekly / monthly) with the selected model's monthly cap, and DeepSeek account balance — each only while a matching model is the active selection.
Agent collaboration · Financial data · Interface extensions · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:itchenshi/dsh-ready-gui#a886d1165b9c44e9887ad9fd6a8993e23449a0c4&path:plugins/dsh-model-surplus
Peak-hour price guard for DeepSeek API tidal pricing: intercept sends during peak hours (Beijing 09:00-12:00 / 14:00-18:00) and schedule them to run at off-peak prices. / DeepSeek 峰谷定价高峰拦截提醒:高峰时段拦截发送,定时到低价时段自动执行。
Agent collaboration · Notifications · Financial data
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:christophersmith2737-commits/OffPeak
Open In: the pack's own cross-platform file-browser launcher for the vncode Web GUI. Forks the shipped @deepseek-ai/dsh-client-ui-open-in-app browser bundle so the File Explorer / Finder / Files menu entries post to a route this package owns, which opens the OS's file browser directly (explorer.exe, open, xdg-open, WSL-aware) instead of the shell's open verb. Alpha.
Agent collaboration · Interface extensions
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:vecnode/vncode#b3a6a0d53b96ef29d452af3962e8d726a122db3a&path:packages/dsh-open-in-app
DeepSeek Harness plugin for the OpenCode / OpenCode Go routes: declares the route wire protocol, keeps the DeepSeek V4.1 models at the front of the model list, and attaches a stable per-conversation x-opencode-session header (fixes 400 MissingSessionID) without leaking the internal session id.
DSH web bundle: inline page previews from `html page-preview` fences plus a floating, session-scoped preview window for local HTML projects and http(s) URLs.
Agent collaboration · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:watericetangcw/dsh-page-preview
PDF as a surface of the pack (row `dsh-pdf`, tab kind `pdf` plus the `pdfs` index page): the agent can actually READ and understand PDFs, SCAN the ones that are pictures of text, and you can read them properly in the right bar. FIVE tools - pdf_info, pdf_read, pdf_find, pdf_render, pdf_scan - are backed by a VENDORED pdf.js 6.3.289 (the same engine and version the shipped preview already uses) with its cMaps, standard fonts AND wasm image decoders, run in a CHILD process with `isEvalSupported:false`, a heap cap and a kill, so a corrupt or hostile document is a reported failure and never a wedged host. Extraction is cached per page under `$DSH_HOME/dsh-pdf/artifacts/<sha256>`, so reading, searching and re-reading a document costs one parse. `pdf_read` offers plain reading order AND a geometric layout reconstruction (lines grouped by baseline, columns and label/value gaps kept), which is what makes an invoice, a statement or a two-column paper legible. `pdf_scan` finds the pages that carry no text layer at all - where recognition is the only way to read them - draws them with the host's own rasterizer (poppler pdftoppm, mutool or Ghostscript) and recognizes them with tesseract, caching each result under every input that can change it (page, language, resolution, segmentation mode). It refuses to OCR a page that already has real text unless asked, and with no engine installed it says which one to install in as many words. The tab type replaces the shipped bare PDF renderer for `*.pdf`: fit-width continuous pages, page navigation, a zoom ladder, a selectable text layer and in-document search; a page that has not been drawn yet reserves the document's OWN page box at the current scale - never a viewport unit, since this reader is one pane of a dock - so the page column can never grow wider than the pane and the fit stays centred while the panel is resized; a THUMBNAIL RAIL and the document's own BOOKMARK OUTLINE in a side panel; every page that has no text layer announces it and offers a one-click scan whose recognized text appears under the page, labelled as a transcription. A second page type (`pdfs`, its own Start-page entry) lists every PDF in the workspace. Read-only: nothing this plugin writes can reach a PDF. No core patches, no fork, no npm dependencies, no network. Alpha.
A playable pixel companion for the DeepSeek Harness Web UI: hatch an egg and pick one of three companions (girl / tabby cat / baby whale), left-click to chat with it through your configured DSH models, platform-jump over chat messages, WASD controllable, with a full nurture system fed by your agent's real token usage.
Agent collaboration · Interface extensions
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:BlackBearCC/dsh-pet-sprite
Cancelable auxiliary-model runtime for DeepSeek Harness with a provenance-preserving usage ledger outside official tokenUsage
Agent collaboration · Developer tools
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Hilbert-beinghappy/dsh-plugin-auxiliary-runtime
Find the right plugin, then get started
Search by task
Use a feature keyword or category, then sort by recent updates or GitHub stars.
Explore capabilities
Read the description, AI analysis and original README to understand what a plugin does.
Check and install
Review compatibility, dependency ranges and instructions, then copy the available install command.
Questions about DSH Hub
What is DSH Hub?
DSH Hub is a community plugin directory for DeepSeek Harness, bringing together descriptions, source links and installation instructions to help you discover and compare extensions.
Does inclusion mean a plugin is tested and compatible?
No. Detail pages show recorded compatibility and dependency ranges. Unknown does not mean compatible; check the project documentation before installing.
How do I install a plugin?
Open its detail page and check the installation section. Copy the command into your Harness environment when available, or follow the project README.