Plugins

Kostenlose Vision für reine Textagenten: integrierte, schlüssellose Vision-Chain plus Pixel-Tools (Q&A, Grounding, Zuschneiden, Pixel-Diff, Farben, OCR, SVG-Trace, Freistellen, Screenshots); Bild einfügen zur Nutzung.

Vision-Aufgaben für reine Textmodelle: intentionsbewusste Bild-Q&A, OCR für lange Screenshots, UI-Reproduktion, Grounding und Pixel-Diff.

Externes Vision-Plugin für DeepSeek Harness: Konfigurationspanel über den Whale-Button, Bilderkennung mit automatischer Antwort sowie Agent-Tools für Screenshot und Erkennung.

HarmonyOS-Geräte-Brücke: hdc-Loop für Screenshot/Installation/Log/Crash/UI-Automatisierung mit read_image, offiziell-vorrangiges versioniertes API-Wissen (SDK .d.ts + gebündelte Offline-Docs) und eine DevEco-CLI-Strecke für Build/Sign/Lint.

DSH↔Godot-Engine-Runtime-Brücke: startet und steuert ein laufendes Godot-4.x-Spiel über dessen In-Game-TCP-Interaktionsserver — 8 Tools (Szenen-/UI-Inspektion, GDScript-Auswertung, Eingabesimulation, Screenshots, Headless-Static-Operationen, Skript-Validierung), ersetzt godot-mcp.

Gemeinsam genutzte echte Browser-Fähigkeit für DeepSeek Harness: ein sichtbarer Browser, den der Mensch beobachten und übernehmen kann, gesteuert vom Agenten über CDP (Snapshot/Ausführung/Inhalt/Tab-Verwaltung); host-seitige Schnittstelle, Provider und browser_*-Tools

Android-GUI-Agent mit ADB-Screenshots, kompaktem UI-Hierarchie-Grounding, verifizierten iterativen Aktionen, Freigaben und einer Mobile-Web-Ansicht.

On-demand vision for text-only DSH sessions: images become markers, and a vision_describe tool sends only image + question to an OpenAI-compatible vision model

Zero-dependency browser control: drive a real headless Edge/Chrome over CDP with 8 `pilot_*` tools (navigate/click/type/keys/eval/screenshot) plus a live draggable cockpit panel in the Web GUI - text-first snapshots for text-only models, no Playwright, no API key.

DeepSeek Harness plugin that turns UI screenshots into implementation-grade web specs using OCR, deterministic geometry, scene graphs, assets, and render comparison.

CloakBrowser-powered native browser tools with per-agent isolated sessions, bounded snapshots, ref-based interaction, screenshots, and safe routing.

Combine text, vision, and image-generation APIs into one Mix model with automatic routing: text-only requests go to the chat model, user images and agent screenshots go to the vision model, follow-ups keep using the same session image, and agents can generate or edit images with session-scoped call history.

Drive the user's real browser through the local Kimi WebBridge daemon: navigate, find-tab, snapshot, click, fill, evaluate, CDP, screenshot, network, upload, PDF export and tab-management tools that reuse existing login sessions.

Playwright MCP browser control for AI agents: drive a real, visible Chrome (navigate/click/type/snapshot + ~80 browser_* tools), with a live screenshot panel embedded in the GUI that refreshes after every action.

Windows computer use for DeepSeek Harness: an MCP stdio server over a PowerShell UIA backend exposing 22 desktop tools (UIA tree, screenshots, typed input, OCR, window management, failsafe).

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

Standalone screen capture for DeepSeek Harness (dsh). Browser hotkeys for instant capture plus an agent-facing capture and read tool that lets the agent see and analyze any screen region.

Browser-Automatisierung für DeepSeek Harness: von Playwright angetriebene Tools open/click/type/screenshot/eval, mit denen der Agent echte Webseiten steuern kann.

DeepSeek-Harness-Plugin: Tools für das Web-Browsing (Markdown / Screenshot / PDF), angetrieben von Cloudflare Browser Run — echtes Headless-Chrome mit JS-Rendering, Login-Sitzungen und WebMCP-Unterstützung.

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.

WeChat Mini Program automated testing for DeepSeek Harness (dsh): selector-addressed actions, build-freshness gates, geometry-first assertions that work on text-only models, and screenshot image blocks on vision routes. 微信小程序自动化测试 dsh 插件：选择器寻址、构建新鲜度门、面向纯文

dsh plugin: launch the dsh-desktop Windows shell (region screenshot, tray, session manager) from the DSH conversation via a desktop_launch tool.

CDP bridge to an already-running Chrome/Edge for DSH: screenshots, pixel assertions, DOM/CSS checks, and page JS evaluation over HTTP + WebSocket, with zero child processes, zero npm dependencies, and zero per-use approval.

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.