Best DSH Vision & OCR Plugins

Browse and search GitHub dsh-plugin repositories from Web UI Settings, sort by stars, and install plugins through the DSH CLI.

Windows-ready DSH computer-use plugin with singleton host dependencies, safe UIA input, and CUA Driver vision fallback.

Cross-platform Computer Use for DeepSeek Harness: virtual-mouse human-like operation (screen_observe + computer_click + type + key + scroll + drag, 11 tools), AX-tree zero-vision-cost mode with free GLM-4V-Flash vision fallback, safety guards (snapshot TTL / app whitelist / dangerous-action approval / password-box protection).

Vision understanding + image generation/editing for DeepSeek Harness (read images with Qwen/OpenAI/Anthropic/Gemini, generate and edit with Qwen-Image). A rework of liustack/modlens (MIT). Remove @liustack/modlens first — both register the same vision providers and image interception hooks, so co-installation breaks image reading.

DeepSeek brain + automatic image transcription: attach images in the GUI and each one is transcribed via the official deepseek-v4-flash-vision-exp by default (a pure-text V4-Pro brain can see images), with any OpenAI-compatible VLM or local Ollama as alternatives.

A vision-language gateway provider route: pasted images are described by a configurable VL model (Qwen-VL by default) before the DeepSeek wire.

Remote host tunnel manager: runs dsh web on a remote Linux server (systemd supervision, no root needed) behind an auto-reconnecting SSH tunnel; allocates and registers remote ports on the server, multi-user safe, with an audit view.

Media and vision workspace for DeepSeek Harness: image, video and speech generation, image Q&A and element locating, long-image OCR, pixel diff, HTML-screenshot verification and video summarization, with DashScope and OpenAI-compatible providers, model pools and a workbench client.

Configurable image recognition for text-only DSH models: image messages are first transcribed by an OpenAI-compatible vision model (Base URL, model ID and API key set in a Settings section) and then passed to the main model as text, while image-input support is advertised. The API key auth scheme is selectable — OpenAI, Anthropic, Gemini or Azure style request headers — and a missing key is reported before any request is sent.

Design-fidelity QA for text-only models: a `deepseek_vision` tool borrows an eye from any OpenAI-compatible vision route, so the model can judge whether an implementation matches its mock — shipped with the benchmark behind that judgement (four fixtures, 23 injected defects, raw transcripts) and the questioning discipline it depends on.

Windows computer use forked from computer-user: 13 computer_* tools that read the screen and drive the mouse and keyboard. Image-capable routes get the screenshot as a real image with an exact image-to-screen mapping, so no external OCR; elements return as UI Automation refs so a click lands on the exact control rectangle; Ctrl+Alt+Esc stops every call.

Read-only Grafana tools that query metrics through the data source proxy: instance health, data source listing, instant and range PromQL queries, current alert state, and provisioned alert rules. Range queries downsample to a point budget; an explicitly supplied step that would exceed it is rejected rather than silently changed, so the model never receives data at a resolution it did not ask for.

Connect a sub2api gateway to DeepSeek Harness: OpenAI-compatible multi-provider routes (OpenAI / Claude / Grok / Gemini) behind one base URL, with per-key model discovery, usage lookup, and global vision/image tools.

Full vision-capability bundle for DeepSeek Harness: a vision_understand tool (OpenAI-compatible vision APIs, free Zhipu GLM-4V-Flash by default) plus paste/drag-and-drop/button entry points for image recognition.

Desktop computer-use for DSH: captures the Windows screen into the agent vision channel and drives mouse/keyboard through a Codex-style overlay with Esc abort; auto-detects Windows native or WSL hosts.

Native vision capability extension, using either Zhipu (free) or Qwen-VL (local).

Vision provider route that transcribes attached images to text through a configurable model (15+ OpenAI-compatible and Anthropic vendors) while DeepSeek keeps answering.

Image-to-text input for the Web UI: paste or drag an image and it is transcribed into structured text and sent, giving text-only LLMs image-input takeover (OpenAI-compatible vision API).

Two-tier image reading for text-only models: fast local OCR (RapidOCR, offline) first, vision-model fallback (modlens).

Vision for any DSH route: paste images in the Web composer with intent-aware auto-analysis, delegate workspace image reads to a Kimi/MiniMax vision subagent, and materialize pasted originals for editing.

OpenAI-completions-compatible adapter for custom gateways (vLLM / LM Studio / self-hosted proxies): always role:"system", thinking driven by the model config, Qwen-style response split, and vision-model image input (single / multiple).

Renders pasted/uploaded images inline in the DSH Web chat and gives text-only models vision: the model-invokable visual_review tool calls any OpenAI-compatible multimodal API first, falling back to a local Qwen3-VL worker.

A content-aware PDF reading plugin for vision models: profiles each page for figures (vector and raster), tables, formula risk and double-column layout, then applies content-aware hybrid extraction, rendering figure/table/formula pages as high-DPI region crops. Provides a low-resolution preview to understand the page layout, and renders a specified region at high resolution. Packaged as multiple tools for agents.

Per-profile settings overlay for DeepSeek Harness: global settings.yaml stays the baseline while each profile overrides any namespace through its own profiles/<name>/settings.patch.yml — object sections merge recursively, scalars and arrays replace wholesale, and !unset masks an inherited value. The overlay is transparent to existing plugins (they keep reading ctx.settings unchanged), writes land in the profile overlay only, and official schema validation, revision semantics, expectedRevision conflicts, watchers and events are untouched. Ships a settings command family (get/set/unset/mask/unmask/promote/demote/migrate/diff/layers) plus a Profile Settings section in the web Settings panel over a loopback RPC channel.

Model-agnostic visual-inspection pipeline for text-only agents: page-by-page HTML screenshots plus a ready-made vision-subagent briefing contract (gewu_prep), then source-code truth verification of every finding (gewu_locate); validated on mimo-v2.5 & qwen3.7-plus.

Windows desktop automation for DeepSeek Harness: accessibility-first observe/act loop with 30 standard tools (element/coordinate targets, auto/a11y/event strategy routing), superseded-observations and no-replay safety semantics, persistent kill switch, RuntimeId/rect-drift stale protection, chat cards and a floating panel (live ops, frame wall, env auto-detect settings), tiled vision describe for low-res models, JSONL audit trail and a pid allowlist. Requires Windows 10+ and Python 3.9+ with uiautomation and pillow (installable from the settings page).

Local OCR using Apple Vision on macOS: text from screenshots, photos, and scans, with no install, no network and no API key.

Windows computer-use capability for DSH: screenshot → vision model → simulated mouse/keyboard with self-evolving knowledge base. Dual-mode plugin tools + skill-only pwsh fallback.

Synesthesia Encoder for DSH: a vision model translates images into compact structured spatial text (canvas/elements/percentage coordinates), giving text-only LLMs pixel-level image understanding via the `mm_vision` tool.

Give your text-only model eyes - chat image attachments are auto-described via a vision model (default prompt), with iterative re-parsing through model-generated prompts when details are missing; system/custom model modes + GUI config panel, key-safe secret handling, and a small host patch for DSH 0.1.0-rc.6 (see repo README).

Local OCR for attached images via Tesseract: only the recognized text is sent to the model, never the image bytes; vision passthrough is opt-in.

Apple on-device Vision tools for text-only dsh models: local OCR (zh-Hans + 30 langs), image classification, face detection, document layout, and a combined describe — 100% offline, no API key, tall-screenshot slicing.

External vision plugin for DeepSeek Harness: whale-button config panel, image recognition with auto-reply, and agent screenshot/recognize tools.

Give text-only DeepSeek models on-demand vision: upload images, DeepSeek answers by calling a view_image tool backed by any OpenAI-compatible vision endpoint (Qwen/DashScope by default).

dsh plugin: model catalog auto-discovery - fetch model listings, pricing and capabilities from OpenAI-compatible API hosts, normalized into ready-to-use config

Windows desktop vision and control for any application that has a window — a native editor, Excel, a proprietary installer, a game launcher — rather than a browser: screen_view returns the live screen as a real image block the model can see (never OCR or a text summary), desktop_control drives the pointer, keyboard and windows inside that same per-monitor-DPI-aware pixel space, and desktop_sequence runs a whole batch of actions in one call — bounded by maxSequenceSteps (24 by default) — so a task costs one round trip and one frame instead of one of each per action.

Gives text-only models vision: forwards user images to an OpenAI-compatible vision model and shows the descriptions in a Web UI right panel.

Vision MCP and DSH bundle for text-only DeepSeek: analyze_image, analyze_clipboard, compare_images and vision_status tools, a visual settings page, free GLM-4.6V-Flash by default, result caching and rate-limit tolerance; keys stay out of logs.

Screen capture and external vision recognition: take_screenshot, list_windows, analyze_image and view_image tools with a configurable GPT vision channel (gpt-5.5 / gpt-5.6-sol / gpt-5.6-terra), API key via the credentials service, and a settings card; view_image shows the screenshot in the Web UI while the model context keeps text only.

Codex-style attachment formats for the DeepSeek Harness Web GUI: PDF text-layer extraction, Office text extraction, scanned-PDF OCR, long-document spill + index cards, image-to-PNG.

Image understanding for any DSH model: vision, OCR, grounding, and crop tools with domain presets for histopathology, cell biology, anatomy, clinical images and scientific figures.

Agent-callable vision tool that describes local images via any OpenAI-compatible vision endpoint you configure, with an optional multi-model cross-check and no built-in keys.

Chemical safety intelligence through the hosted MSDS Chain MCP endpoint: 23 tools for compatibility and mixing order, GHS hazards, PPE, storage, waste disposal, exposure limits, transport classification, multi-region regulatory compliance, SDS lookup and version diffs, and signed audit reports, each answer citing the supplier SDS and revision date it is grounded in.

Auxiliary models for DeepSeek Harness: vision understanding and context compression through dedicated model routes. DeepSeek Harness 辅助模型插件：为视觉理解、上下文压缩、审批审查、子代理、会话标题与图片生成提供独立的模型路由、工具与系统提示，全程不触碰主对话模型。

Integrates TypeSafe Jev for skill and file selection, task supervision, shared-finding corrections, tool-output filtering, approval assistance, and recorded-stage navigation in DeepSeek Harness; all features are disabled by default.

dsh-autovision: paste an image into a text-only model composer and a configured multimodal model transcribes it to text automatically. Twin-provider auto-routing + agent-callable read-image tool. No built-in keys, no relay.

ROS2 debugging toolset and robot-state vision analysis for DeepSeek Harness: node/topic/service/action/interface/TF enumeration, whole-graph topology JSON, rosdep checks, approval-gated builds and custom message scaffolding, GUI screenshots and multimodal vision observation, plus headless RViz2 offscreen rendering (low-poly meshes, direct pixel read, GPU passthrough - motion rendering at 30Hz) with parallel VLM realtime analysis.

An editable diagram canvas in the DSH sidebar plus two agent tools (diagram_read, diagram_apply) that read and edit the workspace in place, using native .drawio files: the canvas opens and saves them losslessly — cells it does not understand are preserved byte-for-byte — and a file-fingerprint revision powers optimistic locking against concurrent edits.

Adds a configurable vision model to text-only main models: a vision_read_image tool, a composer-bar vision-model selector, and automatic image-to-text conversion for text-only routes.

Vision-augmented DeepSeek adapter plugin for DeepSeek Harness: a vision-capable model describes image input, then a text-only DeepSeek model reasons over the description

Computer use through an external USB HID device: name a target in plain language, a vision model locates it in a cropped window screenshot, and a Raspberry Pi Pico 2 moves the real mouse and types on the real keyboard. 7 tools. Windows has window listing, per-window cropping and automatic window raising; macOS needs pyobjc and has neither; Linux is unsupported.

Codex-backed `web_search`, `image_gen`, and `image_vision` tools for DeepSeek Harness, reusing ChatGPT OAuth login state.

Fully-local image understanding & OCR via macOS Vision Framework: `ocr_image` (text, table layout + coordinates) and `view_image` (scene, faces, QR) — paste multiple images into the web input box or pass path/URL/base64; images never leave your Mac.

Local OCR for attached images via the built-in Windows engine (Windows.Media.Ocr): only the recognized text is sent to the model, never the image bytes; vision passthrough is opt-in.

Advertise image paste on text-only DeepSeek routes, describe attachments with a vision sidecar, and leave native vision models untouched.

Bilibili video analysis: metadata, transcript (ASR fallback via Bijian/sherpa-onnx/whisper.cpp), comments, danmaku, and sharp keyframes with optional local vision descriptions.

Registers a `deepseek-vision` provider route: the Web GUI accepts pasted images and transcribes them to text via the free Zhipu GLM vision API before delegating to the DeepSeek adapter.

Drag-and-drop file upload for PDF, Word, Excel and images: dropped files are saved to a local directory and referenced by path, no base64 bloat in the chat.

Headless-browser verification tools so the agent opens the page it just built, reads the rendered DOM and computed styles, checks the console, and screenshots the result; ships a frontend-verify skill.

Local eyes for text-only models: eyes_render draws text/shapes/Mermaid onto a canvas in the Web GUI, eyes_paste captures pasted images, eyes_ocr reads text via the built-in Windows OCR (offline), and eyes_analyze inspects pixels as structured data — no vision model required.

Model-facing `vision` tool for DeepSeek Harness: describe and OCR image files by calling the free Zhipu GLM vision API directly (glm-4v-flash fallback chain), no external CLI required.

Vision through NVIDIA NIM: mounts the bundled stdio MCP server as mcp__opencode_eyes_nvidia__describe_image, which describes an image with any of seven hosted vision models (MiniMax-M3 by default) and rotates several API keys.

Give DSH text-only models vision: an image/OCR/document recognition skill (race pool → custom channels → local) plus an idempotent host patch so image messages reach the model.

Adds a settings page for declaring per-model reasoning efforts and vision support on pi-ai provider routes.

Auto-switch DeepSeek models: routes every request to the vision model when images are present, to the pro model for complex tasks, and to the fast model otherwise; includes a switch_model tool for manual overrides.

Eyes for text-only DeepSeek Harness agents: built-in free vision chain (no key) + pixel-level vision tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots). One-command install, no Python, image turns work like ordinary tool-calling turns.

Gives a text-only model image understanding via a vision-language model, exposed as a `describe_image` tool.

Transparent multimodal routing for text-only models: every image in every model call is fully transcribed (verbatim OCR, data, uncertainty zones, injection-hardened) by your own multimodal understander, with focused re-look via vision_relook and automatic retry on image-related failures. No bundled endpoints, no borrowed logins.

For the DeepSeek Harness native vision model deepseek-v4-flash-vision-exp, raises image admission limits to 32 MiB / 8192 px / 600 images and adds a highres_read tool that tiles large images, then returns the whole image plus 800x800 tiles through the host read_image tool.

Render and iterate on Blender scenes from DSH: 16 tools for scene specs, previews, visual review and delivery renders, with immutable revisions and an approval gate.

Provides the see_image tool for DSH: send an image file to a configurable OpenAI-compatible vision model and relay its description back to a text-only model.

Lets text-only models handle pasted chat images, with a native vision experience, batch image viewing, and a built-in OpenAI-compatible analyze_image tool; vision-capable models are unaffected.

Aura Vision - free vision OCR plugin for DeepSeek Harness web profile (permanent bundle, GLM-4V-Flash free tier, tile-based long-document recognition)

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.

MCP server that lets an AI agent see and operate an Android device you own over adb, for app testing and automation. Vision plus UI-tree fusion, for any MCP client.

OpenCode Go subscription chat for DSH with per-model Completions, Responses, or Anthropic protocol selection, model discovery, context, vision and thinking metadata, effort defaults, and usage meters.