Plugins

Read a video with a text-only model: with ffmpeg installed, sample frames at a 0.1-10 second interval into contact sheets where every frame carries its real timecode, extract the audio track and send it to iTingnao (needs an ITINGNAO_API_KEY) for a timestamped transcript, and describe each sheet with an image-capable model in the host, so the plugin works on its own without any other vision plugin. A composer slider sets the sampling interval, and pure audio files are transcribed without frame sampling.

Adds an ego-webqa MCP server that audits deployed pages across the mobile viewport matrix and returns structured repair evidence; needs uv/uvx on PATH and a Playwright Chromium headless runtime on the same machine.

Compile a video link into timestamped Markdown knowledge cards with back-links to the exact moment. Requires your own Python 3.10+ and yt-dlp (ffmpeg + faster-whisper only for videos without subtitles; Bilibili login cookies for members-only subtitles).

Rejects foreground waiting in DSH agents: waiting commands (sleep, wait, tail -f, watch, ping without -c, infinite loops), blocking wait tools (wait_agent, job_output with wait: true), progress polling (the same check command repeated within a window) and commands that need human input/confirmation (sudo without -n, ssh without BatchMode, login flows, interactive editors, docker -it) are refused with an actionable retry to run in the background or end the turn. Configurable budget and polling window, message language follows the system (zh/en), zero runtime dependencies.

3D director desk sidebar plugin for DSH: stage characters, set camera moves and record timeline previews; AI agents drive the scene in real time via MCP tools.

Take a breath mid-task: the current step finishes, then it holds. Resume sends no new request and spends no tokens; the timer freezes. Current session only.

Timed send for the chat composer: pick a moment and the host delivers the message then. Inside an existing conversation tasks join that conversation; from the new-session screen each task gets its own pre-created session.

Adds an Auto-rename row to each Session sidebar menu: the model reads the Session log, weights recent user messages higher (exponential decay, 3-day time constant), and writes a title of at most 10 CJK characters into the Session title. Also opens that Session menu on right-click, and exposes the same operation as the session_auto_rename agent tool. Requires DSH 0.2.1-rc.5 or newer.

Adds per-model hide/show switches for custom (llm-pi-ai) providers to the DSH settings page. Hidden models disappear from the conversation model picker and are archived verbatim, so they can be restored at any time; disabling or uninstalling the plugin restores everything automatically. Built-in DeepSeek providers are not managed.

Adds the models of your own GPU Hugging Face Space to the model selector, starts the Space when you send a message, shows its start progress, and pauses it when idle or at a time limit. Requires a Hugging Face account with a GPU Space serving an OpenAI-compatible API, a Hugging Face token stored in the DSH vault as HF_TOKEN, and a paid plan for GPU hardware; Hugging Face bills the GPU time.

One-line notice at the top edge of the DSH window with your DeepSeek balance, the ChatGPT/Codex five-hour and weekly remaining quota and their reset times, and how many quota-reset cards are left plus the nearest expiry; the reset-card part links to the ChatGPT usage page in your browser.

Shows the live DeepSeek account balance in the composer toolbar and prices each completed turn in CNY beside its token usage, using the official peak/off-peak table: peak on Beijing-time weekday mornings and afternoons, valley on weekends and statutory holidays. The balance needs a DEEPSEEK_API_KEY credential; accounts signed in through DeepSeek OAuth show no chip.

Windows only (needs the Edge WebView2 Runtime): pop the current session into an always-on-top WebView2 mini window with a self-drawn title bar, free resizing and a jump back to the main window; adds full-screen and region screenshots that paste straight into the composer, a "ask about selection" pill for text selected in any application, global hotkeys and a settings section.

A frosted-glass status card that floats on the Windows desktop and shows what DSH is doing — running tasks with elapsed time and current activity, pending approvals, today's token usage and the countdown to the off-peak window — and collapses into a 14×120 bar that stays out of the way.

A draggable floating clock for the DSH Web GUI with analog and digital faces, a focus timer, four themes, and persistent placement.

A permanent status card in the top-right corner of the DSH content area: one line for where this turn's time went (model, tools, waiting), plus collapsible to-dos with a progress bar, the session goal (edit/pause/resume/clear), submitted plans that open in the preview pane, the files this session changed, subagents with their token and duration figures, and this session's background jobs: running rows carry a two-step stop button, finished ones fold into a count line that stays collapsed until opened; it yields space to the content as the area narrows and folds into a round button at the limit.

Work todo reminders in a sidebar panel with four stage tabs (not started, in progress, overdue, done) plus separate recurring and paused filters, priority derived from the end time in real time, pinning that survives overdue, and enable/disable for recurring tasks. Recurrence covers workdays (following China's statutory holidays and make-up workdays), selected weekdays, monthly dates with month-end rollover, and 5-field cron; one-off tasks default to start and end reminders (start optional, end required); meetings can disable the end reminder. Alerts appear in a page overlay with optional browser notifications, snooze and quiet hours. Twelve agent tools; weekly/monthly reports, natural-language quick add, and cross-view search; data is a single local JSON file with completed tasks archived per quarter.

Tints the strip above the composer green or red to match DeepSeek's off-peak and peak billing windows, shows the remaining time of the current window as a progress bar, and ships six themes plus a custom-CSS hook.

Button beside the composer that rewrites the current draft into a more executable prompt in place. Ships four built-in presets spanning light polish to aggressive expansion, accepts user-defined presets, and switches preset, prompt model and reasoning effort directly from the button without leaving the composer. The button appears only when the draft has content, shows a running timer while enhancing, and surfaces failures with a retry.

Read-only Windows AppData analyzer: click the sidebar icon to drill down explorer-style through directories, each row showing recursive size, file count and last-modified time, plus the software that created it and a delete-worthiness verdict built from a knowledge base, mixed-directory splitting, then heuristics, with unmatched guesses marked as such. Scans directories only and never reads file contents; it ships no delete action, and unknown directories can be looked up online with a model summary and cited sources. The online lookup needs a DeepSeek API key (or an equivalent credential); without one the query is only copied to the clipboard.

A Windows pixel Madara desktop companion with 16 animation clips driven by Harness task events, using the original BigFish runtime dependency.

Give a text-only agent video understanding: extract timestamped frame files, sample only on scene changes, export GIF clips, probe metadata, and run a one-shot analyse over the frames. Five tools that shell out to a local ffmpeg through argv arrays rather than a shell string.

Local video color editing with camera-matched DJI Log LUTs, independent highlights and shadows, optional skin and selective color controls, frame previews, and 10-bit HEVC export. The universal tarball listed here ships no runtime: its first-launch "configure environment" button downloads a pinned platform runtime (Python 3.12 + FFmpeg 9, ~100 MB) over HTTPS, or point it at your own Python 3.10+ and FFmpeg/ffprobe 8+; the platform-suffixed release packages bundle that runtime.