Voice input for the dsh web GUI: a mic button in the composer tool row that transcribes speech into the draft via the browser Web Speech API (zh-CN by default).
Terminal-style input history recall for the DeepSeek Harness web composer: bare Up/Down move the caret first, only the draft edges trigger recall, the in-progress draft is stashed before browsing and restored exactly when you return to the newest entry or press Escape.
Thin DSH bundle that bridges design-playbook's preview + evidence MCP servers into a DeepSeek Harness profile. Depends on the design-playbook package for the server implementations.
Agent 协作 · 开发工具 · integrations · 界面 · skills
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Bandersnatch0x/design-playbook#33c1298d0eb5c0f0564c29235fcaf612e0a94c47&path:packages/dsh-design-playbook
Codex-style drag-drop for DSH web GUI: drag any file, it lands in ~/.dsh-dropbox, the absolute path text is inserted into the composer input. Hot-pluggable: install with `dsh plugin --profile web add ./plugins/drop-file-to-path`, no dsh source changes.
Agent 协作 · 界面
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:GLFzr/dsh-drop-file-to-path
Every second there is a 1-in-1000 chance that Foxy (Withered Foxy) from Five Nights at Freddy's 2 jumpscares you in the DeepSeek Harness web UI: a full-screen scare plus the iconic scream.
界面
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:01Virex/dsh-foxy-jumpscare
LAN access for DeepSeek Harness Web UI: loopback-preserving reverse proxy with Host/Origin rewrite, in-app browse directory picker, and a secure-context UUID polyfill. For trusted LANs only — the harness web control plane has no authentication.
MCP server manager for DeepSeek Harness: manage MCP servers from the Web settings page, stored globally in $DSH_HOME/mcp-servers.json, connected/disconnected at runtime.
DSH Web 麦克风语音输入插件:浏览器内置 Web Speech API 实时转写进输入框,自动去重/续听、智能标点、语言与自动发送设置(Edge=微软语音、Chrome=谷歌语音)。Microphone voice input for the DSH Web UI using the browser's Web Speech API.
界面
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:QT-Chen/dsh-mic-input
DeepSeek Harness plugin: one-page mobile access setup — Tailscale/LAN detection, scan-to-open QR (offline, server-side), and an insecure-origin crypto.randomUUID polyfill so the web UI also runs on plain-HTTP phone origins.
integrations · 界面
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:YiYan129600/dsh-mobile-access
Multimodal eyes and hands for DeepSeek Harness: GLM vision transcription, OCR, and text-to-image via DashScope / OpenAI-compatible backends, with an in-conversation generated-image card.
界面 · 效率 · 视觉
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:MC5lan/dsh-multimodal