Vision &amp; Media DSH plugins

@lamplitisles/kepos-speech

Kepos Speech

Kepos Speech plugin for Alibaba and ByteDance Chinese speech synthesis plus short-audio recognition in DeepSeek Harness Web

Voice input plugin for DSH: tap or hold Alt to talk, then release or tap again to convert speech to text. Supports a hotword replacement list (hot.txt), custom polishing prompts, and a recording level indicator. Uses local offline recognition by default (SenseVoice; zero configuration, zero key, and

Image Viewer

Zoom, pan, download, gallery, and region-note image viewer for DeepSeek Harness

Live2d Avatar

A switchable Live2D avatar stage and desktop companion for DeepSeek Harness.

ui-customizationvision-media★ 0↓ 0

v0.3.1

dsh-grok-adaptation

Grok Adaptation

Normalize undersized Grok Responses images with Sharp while leaving supported large images unchanged.

vision-mediaintegrations-communication★ 0↓ 0

v1.0.0

dsh-vision-pro-bridge

Vision Pro Bridge

Give text-only DeepSeek-V4-Pro real vision with zero new dependencies and DeepSeek-only routing: images are described by deepseek-v4-flash-vision-exp (your existing DEEPSEEK_API_KEY), then the text is handed to V4-Pro.

vision-mediamodels-usage★ 0↓ 0

v1.0.2

dsh-vision-toggle

Vision Toggle

DSH model vision toggle: adds a “Model Vision” row to the settings page for switching the input vision modality of manually declared models under llm-pi-ai custom routes via the official settings.mutate channel, with changes taking effect immediately.

v0.2.10

dsh-asr-voice

Asr Voice

Speak and it becomes text · Speak-to-prompt for DeepSeek Harness: cloud ASR speech recognition + prompt optimization + fill into draft/auto-send, cross-platform macOS / Windows.

@dshp-inx/vision-bridge

v1.3.0

Vision Bridge

DeepSeek Harness (DSH) vision bridge: let text-only models delegate image understanding to multimodal vision models via vision_describe tool, with persistent settings UI and primary/fallback auto-retry. 纯文本模型通过 vision_describe 工具把图片转交给多模态视觉模型识别的桥接插件。

vision-mediaintegrations-communicationmodels-usage★ 0↓ 0

productivity-workflowvision-media★ 0↓ 0

v1.1.0

dsh-pdf-to-word

Pdf To Word

DeepSeek Harness plugin: PDF→Word (.docx) conversion with layout fidelity (fonts/tables/images/borders), OCR scan mode, and optional multimodal LLM verification. Registers the pdf_to_word model tool.

vision-mediamodels-usage★ 0↓ 0

dsh-vision-patch

Vision Patch

Per-model and per-route image-input (vision) checkboxes on the Models page's custom-provider cards, writing through the llm-pi-ai settings namespace.

dsh-plugin-vision

Plugin Vision

Give the DSH agent a pair of eyes: call online VLMs (multi-provider, OpenAI-compatible) to analyze local images, URLs, and attachments uploaded in conversations; the selected models’ specialized capabilities are injected into the system prompt in real time

Imgdraw

Text-to-image for DeepSeek Harness: a `draw_image` model tool, an input-bar 生图 button with a prompt popup (async generation, 4-grid results, download / keep / delete), an /imgdraw image route, and persisted history. Backends: DashScope wan2.7-image (free default) and SiliconFlow Qwen-Image.

vision-mediaagents-orchestration★ 0↓ 0

v0.1.2

dsh-vision-autoswitch

Vision Autoswitch

DeepSeek Harness plugin: auto-route image-bearing requests to deepseek-v4-flash-vision-exp, then fall back to the original model.

Updated 9/2/2026

v1.0.0

dsh-vision-assist

Vision Assist

DeepSeek Harness vision assistant plugin: equips models without vision capabilities with a switchable multimodal recognition model. Images in the input field are automatically saved to disk and rewritten as text prompts; the main model can view images by calling the vision_recognize tool. The recogn

vision-mediamodels-usage★ 0↓ 0

Updated 9/2/2026

v0.3.9

dsh-vision-bridge

Vision Bridgealt · alaxrpg

DSH Vision Bridge Plugin: Reuse DSH Provider or connect directly via OpenAI compatibility, with visual configuration and image pasting support

vision-mediaintegrations-communication★ 0↓ 0

Updated 9/2/2026

vision-mediaui-customization★ 0↓ 0

dsh-plugin-show-image

Plugin Show Image

Render local image files inline in the DSH conversation via a global show_image tool.

Media Preview

DSH 插件：在聊天记录中自动将本地音视频/图片路径渲染为可播放的预览组件。When an assistant message or tool result contains a local media path, the path is replaced inline with a playable <audio>/<video>/<img> element backed by same-origin /api/media-preview/* route.

Updated 9/1/2026

v0.2.5

dsh-plugin-voice

Plugin Voice

DeepSeek Harness plugin: voice + notification outputs—agents proactively contact users through cloud TTS (Volcano seed-tts / Xiaomi MiMo V2.5, with automatic fallback to SAPI on failure), desktop notifications, and alert sounds. Combines the native DSH integration of dsh-plugin-notify with the cloud

vision-mediaintegrations-communication★ 0↓ 0

Updated 9/1/2026

v0.1.3

@maiziman/dsh-model-capabilities

Model Capabilities

Automatic reasoning and image capability detection for custom DeepSeek Harness models

Llm Multimodal

DSH plugin: Provides image/video generation tools in DSH, based on OpenAI-compatible APIs. Models are automatically discovered from existing llm-pi-ai settings.

@xiaokaizhou/dsh-media-preview

v0.1.1

Media Previewalt · @xiaokaizhou

vision-mediaui-customization★ 0↓ 0

memory-contextvision-media★ 0↓ 0

v1.0.3

dsh-image-amnesia

Image Amnesia

Keep native vision, drop historical images before they hit relay providers. Global DeepSeek Harness bundle for every agent.

v1.0.0

dsh-voice-control

Voice Control

Voice control for DSH web: speech-to-text into the composer (with auto-send) and spoken playback of assistant replies via the Web Speech API

Photos

Photo upload for the DeepSeek Harness composer — native picker feeding the stock paste pipeline

@jetecho/dsh-csv-and-image-preview

v0.1.2

Csv And Image Preview

Preview images / SVG and CSV tables in the DeepSeek Harness chat, rendered as real browser <img> / <table> elements. Preview-first workflow: show the user the asset, wait for approval, then apply the real change.

vision-mediaui-customization★ 0↓ 0