CATEGORY
Verified manifests and exact versions in this category.
dsh-llm-multimodalDSH 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-previewDSH 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.
dsh-image-amnesiaKeep native vision, drop historical images before they hit relay providers. Global DeepSeek Harness bundle for every agent.
dsh-voice-controlVoice control for DSH web: speech-to-text into the composer (with auto-send) and spoken playback of assistant replies via the Web Speech API
dsh-photosPhoto upload for the DeepSeek Harness composer — native picker feeding the stock paste pipeline
dsh-multimodal-runtimeMogu Multimodal Runtime - DeepSeek Harness 统一多模态能力运行时 (Comfy Local Provider V1)
dsh-speech-inputA microphone button for DeepSeek Harness that writes browser speech recognition into the composer draft.
dsh-maclensBridge Apple's on-device Vision framework (macOS) into DeepSeek Harness: OCR, image classification, face detection, and document layout as local dsh tools. No network, no API key, no daemon.
dsh-tu4-inline-imagesInline images in conversations DSH plugin — renders local image paths in the input box, user messages, and LLM replies directly as images (loopback routing + strong token + multiple-root allowlist + three size presets). Inline local images in DSH web conversations.
dsh-model-auto-hot-switchAutomatic per-task model hot-switching for DeepSeek Harness (dsh): image-aware tasks route to the vision model automatically, every other task keeps your default model. Zero extra tokens, no context disturbance.
xby-obb检测图像中的旋转目标，输出旋转边界框、角度、置信度和类别标签。支持15个目标类别：plane， ship，storage tank，baseball diamond，tennis court，basketball court，ground track field， harbor， bridge，large vehicle，small vehicle， helicopter， roundabout，soccer ball field，swimming pool。
xby-ocr-bank-cardIdentify bank card numbers, issuing banks, and card types; validate card number validity using the Luhn algorithm.
xby-ocr-captchaSubmit a common CAPTCHA image and return its text.
xby-ocr-driver-licenseRecognizes the main page of a driver's license (license number, name, gender, nationality, address, date of birth, permitted vehicle classes, date of first issuance, validity period) and the secondary page (file number).
xby-ocr-handwritingInput an image containing handwritten text to automatically detect text lines and recognize their content. Suitable for handwritten notes, signatures, handwritten forms, and more.
xby-ocr-id-cardRecognizes the front of an ID card (name, gender, ethnicity, date of birth, address, and ID number) and the back (issuing authority and validity period), automatically determines the front or back, and validates the ID number.
xby-ocr-passRecognizes the permit number, name, gender, date of birth, validity period, place of issue, and other information on Exit-Entry Permits for Travelling to and from Hong Kong and Macao and Taiwan Travel Permits, with support for MRZ parsing.
xby-ocr-passportRecognizes passport number, Chinese name, English name, gender, nationality, date of birth, date of issue, expiry date, place of issue, and other information; supports MRZ parsing.
xby-ocr-proHigh-precision text recognition. Input an image containing text to automatically detect and recognize its content. Suitable for various scenarios, including documents, billboards, and screenshots.
xby-ocr-vehicle-licenseRecognizes the license plate number, vehicle type, owner, address, make and model, engine number, vehicle identification number, and other information on motor vehicle registration certificates. Supports automatic orientation detection and filtering of primary and secondary pages.
xby-ocrText recognition balancing speed and accuracy. Automatically detects and recognizes content in images containing text. Suitable for various documents, billboards, screenshots, and other scenarios.
xby-pedestrianInput an image, detect pedestrians in the image, and output the bounding boxes, confidence scores, and labels for all detected objects.
xby-pet-detectRecognize pet facial expressions and output 4 emotion categories: Angry / Happy / Relaxed / Sad.
xby-picIncludes general text recognition, handwriting recognition, license plate recognition, ID card recognition, everyday object detection, insect recognition, plant recognition, passport recognition, Hong Kong, Macao, and Taiwan travel permit recognition, bank card recognition, business license recognit
xby-plant-recognitionIdentify the plant name (or its family, genus, species, or subspecies).
xby-plate-recognitionRecognize license plate numbers, license plate colors, single- and double-layer license plates, and bounding boxes.
xby-poseDetect people in images and output bounding boxes and keypoint coordinates. Each person has 17 keypoints, with each point representing a different body part, in order: nose, left eye, right eye, left ear, right ear, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, lef
xby-read-pdfAn MCP server that enables AI assistants to read and analyze PDF files, providing PDF metadata extraction, page-range reading, and keyword search.
xby-recogIncludes general text recognition, handwriting recognition, license plate recognition, ID card recognition, passport recognition, Mainland Travel Permits for Hong Kong, Macao, and Taiwan, bank card recognition, business license recognition, driver's license recognition, and vehicle registration cert
xby-reflective-vestInput an image, detect whether personnel are wearing reflective clothing, and output bounding boxes, confidence scores, and labels (safe/unsafe) for all targets in the image.
