Плагины

Adds a path-backed attachment dock for images, documents, archives, code, and other files, leaving file reading to the Agent's available tools.

Adds ChatGPT/Codex OAuth-backed Codex models and gpt-image-2 generation/editing to DSH without requiring an OpenAI Platform API key.

GitHub Copilot LLM adapter: OAuth device-flow sign-in, live model discovery from the Copilot API, vision support for image-capable models (gpt-4.1, gpt-4o), and two wire protocols (Chat Completions and Responses API) with automatic endpoint routing per model.

Persistent wallpaper skin for the dsh web profile: static image or muted looping video as the app background, with translucent panels and a floating picker panel.

Workspace file tree for the Web UI: multi-tab preview (text/DOCX/PDF/Markdown/CSV/images), inline Markdown editing, a Git changes panel, and file operations (rename/delete/copy/move/create) via context menu.

Скин с обоями для веб-интерфейса DSH: фоновое изображение с настройкой прозрачности, маски и размытия.

Улучшенная версия dsh-skin: переключение скинов и собственные фоновые обои для DeepSeek Harness. По сравнению с оригиналом исправлены две проблемы — ① зависание при загрузке крупных изображений/нестандартных форматов (теперь декодирование напрямую через createImageBitmap, добавлены индикаторы обработки и ошибок); ② переполнение стека из-за бесконечной рекурсии при смене темы, когда overrideTokens вызывал theme/change (добавлена защита от циклов при смене скина/светлой-тёмной темы). Хранение обоев переведено с сжатого варианта в localStorage на прямое хранение исходного файла в IndexedDB — полное качество изображения, нулевое сжатие, с автоматической миграцией старых данных.

Плагин внешней vision-модели для DeepSeek Harness: inspect_image отправляет локальное изображение или URL http(s) на любой совместимый с OpenAI эндпоинт, а текстовый ответ vision-модели по картинке возвращается прямо в диалог; с панелью настроек в веб-интерфейсе.

Детерминированная маршрутизация по ключевым словам, переключение моделей по allow-листу и изолированный канал image_gen для DeepSeek Harness.

Per-file-type multimodal chains: preset processing models per wildcard convert image/video/audio files into prompt tokens before they reach the text-only session model, with per-preset fallback chains and a Multimodal settings page.

DeepSeek Harness(DSH) web plugin: paste files into the composer, inbox storage, thumbnails. Text via read_inbox; image understanding needs your vision skill (bundled image-reader + ARK_API_KEY).

A Genshin Impact Columbina (哥伦比娅) themed plugin for DeepSeek Harness Web: preset/custom background images with a translucent mask, per light/dark scheme palettes, and a special glow for the active conversation title.

DeepSeek Harness plugin: a `deepseek-vision` route that declares image input and transcribes pasted images via the free Zhipu GLM vision models before delegating to the DeepSeek adapter.

Multimodal plugin for DeepSeek Harness: understand images and generate images through configurable OpenAI-compatible or DashScope endpoints.

Paper Design MCP bridge as a DeepSeek Harness bundle: paper_* tools, image screenshot support, design skills, OAuth auth, and Cursor-parity design guidance.

Wallpaper Engine integration for the dsh web GUI: scan the local Wallpaper Engine library (Steam workshop 431960 + local projects), use its wallpapers as the page background (image / video / web / scene-preview), and control them from a right-side '婢逛胶鐒婄拋

DeepSeek Harness out-of-tree plugin: give a text-only coding model eyes by routing images to a Qwen-VL (DashScope) route through ctx.llm and returning text.

Custom background (images / GIFs / animated wallpapers) plugin for the DeepSeek Harness (DSH) Web GUI: an enable switch plus a local asset library, with separate display mechanisms for the official theme and skin-center skins.

DeepSeek Harness plugin: let text-only models receive pasted images, and analyze them with a built-in OpenAI-compatible vision tool

Google Antigravity CLI (agy) as a DSH product subagent: delegate tasks to a Gemini coding agent with your Antigravity login (no API key), with conversation follow-ups, background streaming progress, image analysis by path, and a pop-up login window.

WeChat Official Account content studio: anti-homogenization rotation writing, low-creativity remediation playbook, blessing-image visual baseline, gpt-image pipeline with OCR acceptance, and the measured xiaolvshu draft web API.

Registers the remove-pdf-background-gray skill for DSH: whitens gray/off-white scan backgrounds in image-based PDFs while preserving resolution, page geometry, and anti-aliased text edges (lossless Flate write-back), via a single Python script (pypdf + Pillow + numpy).

Chat image-attachment bridge with a `view_image` tool for any OpenAI-compatible VLM (local Ollama or cloud): pasted/dropped images become `view_image` path markers before reaching text-only DeepSeek models.

`describe_image` tool: a vision bridge that sends images to mimo-v2.5 through the opencode Zen API (credential `OPENCODE_GO_API_KEY`, free route first with paid fallback) and returns text descriptions for text-only models, with native passthrough and ImageMagick transcoding of SVG/TIFF/HEIC formats.