Плагины

Позволяет веб-интерфейсу DSH прикреплять файлы любого типа как @-упоминания без пути: выберите файл рабочего пространства в меню @ или перетащите его в компоновщик, и агент прочитает его напрямую.

Закрепляемый предпросмотр Markdown для DSH Web UI: рядом с беседой, список недавно прочитанных .md для каждого сеанса, локальные изображения по относительным путям отображаются на месте, а нажатие на ссылку .md в чате открывает предпросмотр вместо системного редактора; следует светлой/тёмной теме оболочки.

Упоминания @file/@folder с областью действия сеанса в поле ввода, с Git-осведомлённой индексацией, управляемыми загрузками перетаскиванием и вставкой, а также проверенными маркерами путей.

Боковая панель древа файлов для DeepSeek Harness: панель файлов с ленивым деревом со значками по расширению, действиями «копировать путь» и «отправить агенту».

Инструмент `generate_image`, обращённый к модели, для текстовых моделей: модель просит картинку на естественном языке, Gemini рисует её через CLI Antigravity, изображение сохраняется в настраиваемый каталог вывода, а инструмент возвращает пути к файлам, которые модель может использовать в работе.

Расширенный обозреватель каталогов рабочего пространства для веб-интерфейса: доступ в один клик к дискам Windows /mnt под WSL, полная цепочка предков и всегда видимое поле ввода пути.

Мост вложений-изображений в чате с инструментом view_image для любой OpenAI-совместимой VLM (локальная Ollama или облако): вставленные/перетащенные изображения становятся маркерами путей view_image, прежде чем достичь текстовых моделей DeepSeek.

Двухвременная (bi-temporal) память на базе демона SodaMem: воспоминания подмешиваются в промпт при сборке каждого хода, и каждый закрытый ход загружается в память — без вызова инструмента и без вызова модели на пути чтения; каждый вспомненный факт ссылается на ход, из которого пришёл, и несёт даты действия, поэтому вытесненные факты перестают извлекаться.

Полноценные инструменты Git для модели: status / diff / log / commit / branch с проверкой сообщений и путей — без прямых вызовов git через shell.

Runs a Douyin subagent that browses videos and reads comments in valley hours and reports back, with dy_* tools for publishing and replying (comments sent from the already-logged-in page with its own request signature, no IM ticket), a daily quota timer, and a settings panel under Settings → Douyin: master switch, auto mode, valley windows, rounds per day and report depth. Windows-oriented; needs the separate douyin-mcp binary (yunsur/douyin-mcp releases, path set via dy_config mcpExe) and an Edge/Chromium login; replying needs a logged-in browser on a CDP port; publishing needs extra credentials most accounts lack.

Hides tool schemas from the model's prefill and reveals one tool at a time through a decision model API. Requires a running decision-model API endpoint and pnpm on PATH.

Five tools that build, test and run the Inimerse engine, and drive its Verse layers over the server's line protocol. Needs a local Inimerse checkout built with CMake; point the bundle's repoRoot at it (the shipped default is the author's path).

Git Bash as the default shell on Windows. Resolves bash.exe from git --exec-path (never the WSL stub), exposes a run_shell tool, and degrades visibly instead of disappearing when Git Bash is missing.

Downloads a Bilibili video with yt-dlp, samples key frames, and transcribes it — from the site subtitles, or from a local faster-whisper install when there are none. Registers the skill the agent uses to write the summary; activates when a user message contains only a video link. Requires yt-dlp and ffmpeg on PATH; transcription needs faster-whisper.

Right-click any file reference in the chat to open it, open it with an associated application, reveal it in the file manager, copy its path or contents, and save it.

Publishes whichever model the profile has selected in chat or on the Models page as an OpenAI-compatible HTTP endpoint: /v1/models, /v1/chat/completions (streaming SSE and non-streaming), /health. A request may name any routable model; one that omits `model` gets the current chat selection, so switching model in chat switches the endpoint. Switched on, it can also lend the model the Harness tools switched on individually, run those calls through the Harness tool runtime and feed results back up to a round limit. The listener is the plugin's own, so 127.0.0.1, 0.0.0.0 and a NAT-forwarded port all behave the same with no change to the request path. Settings cover bind address, port, API key with a generator, fallback provider and model, reasoning effort, CORS origin and timeout.

Converts Office and PDF files in the DSH workspace to Markdown locally — no network, no API quota — before the model reads them. Writes each .md next to its source so only a file path enters the context, and returns byte size, line count, token estimate and a heading outline, for a single file or a whole directory.

Local art pipeline for DSH: drives the locally installed SD WebUI and ComfyUI backends through ten image_* tools and an Art Studio sidebar panel with an embedded ComfyUI canvas, post-processes with the Python shipped with those installs (rembg cutout, Real-ESRGAN upscale, seamless tiling), and writes Godot-ready assets with res:// paths once a project root is set.

Registers one `ocr` tool for text-only models: it runs the offline script bundled in the repo over a single image or PDF and returns the recognized text — Windows native OCR first, then RapidOCR, then Tesseract. Recognition runs locally with no network request and no API key. Needs Python 3.11 on PATH plus `pip install winocr pillow` (winocr is the preferred engine; RapidOCR and Tesseract are fallbacks). Per-line results for images, per-page results for PDFs.

In-session cd for DeepSeek Harness: re-base relative paths on the fly (the file tools and the bash/pwsh default workdir follow) or move the session to another workspace for real — a `cd {"workspace": true, "confirmed": true}` call the user has agreed to, or the user's own `/cd <directory>` command, relocates the session's stored log, sandbox root, AGENTS.md and skill discovery and sidebar grouping. Per-session, persisted across restarts, without moving the session for a plain re-base.

Two-tab archive panel (archived / active) where every row offers archive or unarchive next to permanent delete; each row shows its bound workspace folder as a clickable path, and queued archive-entry cleanup keeps a deleted session from resurfacing in the active list.

Enhanced right-sidebar files panel: rename, create, upload, download (a directory is zipped on the fly), copy the full path and delete from the tree; an editable document renderer for 128 text extensions with Ctrl/Cmd+S save and conflict detection; plus an independent bottom terminal panel with tabs and a quick-command library persisted under the DSH home.

Composer prompt enhancement for DeepSeek Harness: a button beside the model selector rewrites the draft through one auxiliary LLM call and writes the result back in place, with composer undo left intact; a /prompt-enhance slash command returns the rewritten prompt as a command result line without touching the draft; the enhancement call can be pinned to its own fast provider and model in the profile patch; the rewrite policy is faithful - original intent, language and information are kept, code, commands and paths are preserved verbatim, and nothing is invented or padded with placeholder text.

MCP manager for DeepSeek Harness Web: a Settings page beside Plugins that lists the MCP servers the running loader tree actually holds, each with its config summary, endpoint and live state, and disables or restores any of them through a paired-marker block in the profile's cordis.patch.yml. Env values, header values and `!!js` sources reach the page only as key names or dot-paths.