vision
Vision DSH bundle: give the agent screen/window vision — see tool captures via the bundled Python cvision and returns the image natively
155 results
Vision DSH bundle: give the agent screen/window vision — see tool captures via the bundled Python cvision and returns the image natively
Repository and agent-runtime analysis skills as an installable DeepSeek Harness plugin: repo recon, DSH enhancement analysis, runtime optimization, publish screening, and DSH plugin-potential assessment.
WeChat-style in-app screenshot button for DeepSeek Harness: drag to capture a screen region, PNG saved with auto-numbering, marker [Shot N HH:mm] written into the composer.
Erii (Sakura) themed boot splash animation for the DeepSeek Harness Web UI: falling sakura petals, a mint monster mascot with a cherry blossom on its head, a soft pink progress bar, and the line "Sakura, walk slower." Auto fades out after ~3 seconds; pure client-side, zero dependencies.
A liquid glass splash loader with whale aura animation, real-time plugin checklist, and resident diagnostics for DeepSeek Harness Web GUI.
The small deterministic operations an agent needs mid-task. JSON/YAML round-trip, JSON Schema validation, text diff, JWT decode, regex testing, SQL dialect transpiling, QR codes, timezone and unit conversion, page screenshots.
A playful slingshot toy for the dsh web GUI: drag to aim, fire a stone, shatter UI elements and watch them tumble off screen — then pop back, good as new.
DeepSeek Harness plugin: publish single-screen content cards to WePre Next straight from a dsh agent session
Media player for the dock file explorer: plays audio (MP3, WAV, OGG, FLAC, M4A, ...) and video (MP4, WebM, MOV, ...) files, with a music player for pure audio and fullscreen playback for video.
Pixel-diff two screenshots into diff.png + a labeled triptych (pixelmatch)
DSH plugin exposing adb-based tools that let the model operate an Android phone (screenshot, UI dump, tap, swipe, input text, press keys, open apps). | 让模型通过 adb 操作安卓手机的 DSH 插件(截图、UI 树、点击、滑动、输入文本、按键、打开应用)。
Read-only browser verification tools for the DeepSeek Harness web GUI: browser_open / browser_mock / browser_assert / browser_screenshot — verify a page (H5/desktop) in ≤4 tool calls with mock interception, DOM assertions, and screenshots that auto-project into the model context.
Native full-screen MCP manager for dsh-TUI
DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, OCR, grounding, UI restoration, pixel diff, Artifacts, and Web UI.
Computer Use for DeepSeek Harness backed by the cua-driver daemon (trycua): accessibility element-level targeting, background-first input delivery, window/desktop screenshots, deterministic verification.
Mobile/touch optimizations for the DSH Web GUI: iOS input-zoom fix, safe-area insets, touch targets, and viewport clamps for stock dialogs on narrow screens.
DSH 服务面板:查看由哪个对话(Agent 会话)启动了哪些服务——PEB 环境块读取 DSH_SESSION_ID 归属(含分离进程),三态健康探测 + HTTP 码,支持礼貌停止、日志实时查看与启动参数重放。Host scans agent-started services with session attribution; client renders a fullscreen table panel with tri-state health, graceful stop, live logs and restart replay.
Native macOS Computer Use (screenshot + mouse/keyboard control) tool for DeepSeek Harness
Full-screen Matrix-style TUI frontend for DeepSeek Harness
Accessibility plugin for the DeepSeek Harness web UI: announces tool calls as they appear and finished assistant messages through an aria-live region so screen readers read them aloud.
自适应 GitHub README 撰写与截图技能:先判断目标类型与截图能力——网页走浏览器截图,无网页的控制台/CLI 走终端截屏(System.Drawing / HTML 终端 / termtosvg),能则询问并自动配图,不能则走无截图模板;强调相对路径防丢图。作为 dsh 插件(cordis skill-bundle)安装后由 host 层注册 readme-writer 技能;SKILL.md 含可被插件商城索引的 name/description/keywords/category/version 元数据。
DeepSeek Harness plugin: analyse images out of band with a vision model — pasted images are digested into text before admission and a describe_image tool covers image paths, all without ever changing the session's model.
Zero-token offline math input for DeepSeek Harness: handwriting recognition, screenshot OCR, LaTeX editor, and inline LaTeX rendering.
Sage GUI Toolkit — Windows desktop control tools (screen layout / annotated screenshot / click / type / key / scroll / wait / window / UIA structured query) via PowerShell + SendInput + UIAutomation, DPI-aware and multi-monitor safe