@local/dsh-parakeet-voice-inputOffline Parakeet voice input for DeepSeek Harness
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:vTRKA/voice-stt-dshDiscover The Best DSH plugins
23037results
Search community plugins for DeepSeek Harness. Compare capabilities, installation instructions and compatibility to find extensions for your AI workflow.
@local/dsh-parakeet-voice-inputOffline Parakeet voice input for DeepSeek Harness
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:vTRKA/voice-stt-dsh@local/dsh-pdfPDF reader, annotations, navigation and pluggable OCR for DeepSeek Harness
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:dingyiliao/dsh-bundle-pdf@local/dsh-peak-reminder峰谷提示器:DeepSeek Harness 高峰时段提醒插件(模式指示灯 / 高峰前提醒 / 余额卡片)
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:lsakira/dsh-peak-reminder@local/dsh-penDrawing canvas for the model: brush tools, screenshots, and a zoomable canvas panel in the right sidebar.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:shine-yu-student/dsh-pen@local/dsh-plugin-data-maskPaste-time data masking for the DeepSeek Harness Web composer: sensitive values are replaced before they ever reach the input box.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:virzs/dsh-plugin-data-mask@local/dsh-plugin-local-prompt-bridgeStock-safe DSH plugin: HTTP tokenize pressure + llama.cpp overflow → CONTEXT_WINDOW_EXCEEDED compact-retry. Community dsh-plugin; does not patch @deepseek-ai packages.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:tianyiming1/dsh-plugin-local-prompt-bridge@local/dsh-plugin-token-billingDeepSeek Harness 的 Token 计费插件:当前会话与全账户的 token 用量、估算费用(官方峰谷价+缓存折扣)、账户余额,右侧用量仪表盘(最近对话、余额曲线、热力图、时段消费、模型花费)。目前自用,欢迎建议与意见。
@local/dsh-plugin-voice-inputVoice input plugin for DeepSeek Harness web: a mic button in the composer that recognizes speech (Web Speech API) and fills the draft, with optional auto-send.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:NewDaNew/dsh-voice-input@local/dsh-preset-dispatchAgent 管理: persistent native agent presets with role, permission and dispatch policy in one place
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:BigRagdollCat/DSH-Preset-Dispatch@local/dsh-prompt-optimizer将原作者 linshenkx 的 prompt-optimizer 移植到 DeepSeek Harness 的第三方插件
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:zhang-jiazhi/dsh-prompt-optimizer@local/dsh-provider-quote对正在使用的模型提供商询价:账户余额 + 模型单价(USD / 1M tokens)
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:miku05231/dsh-provider-quote@local/dsh-reading-mode阅读模式:英语文章批注阅读面板(host 分析服务 + client 面板),仅在与该预设同名的会话中渲染
@local/dsh-response-notificationDeepSeek Harness Web 浏览器后台最终响应完成通知
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:sd1g1/dsh-response-notification@local/dsh-review-mode审核模式:明确选择目标、按新用户消息自动复审,2~4条可展开洞察,独立面板不影响正常问答。
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:hqa-shu/dsh-review-mode@local/dsh-rewind-retryRetry, edit-and-resend, or continue any finished turn. Rewinds by forking the session right before the chosen prompt.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Locietta/dsh-rewind-retry@local/dsh-rtl-chatRight-to-left reading mode for the DeepSeek Harness conversation answer area, with a one-click LTR/RTL toggle.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:AdilAlmahmoud/dsh-RTL-chat@local/dsh-secretaryLets one secretary agent create projects (workspaces) and run threads (sessions) inside them.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:LiBinDeve/dsh-secretary@local/dsh-secrets-managerProject secret management for the DeepSeek Harness: discover every .env file across a monorepo (root and workspace packages), edit keys safely in a Settings page, and hand the model a DSH_ENV_FILE / DSH_ENV_FILES pointer so shell commands can source the right environment.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:lifecoder1988/dsh-secrets-manager@local/dsh-session-costPer-conversation token and money accounting: session_cost / list_session_costs tools, the /cost command, and a composer cost pill
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:miaoxingyunduo/dsh-session-cost@local/dsh-session-dustbinDSH session dustbin: delete sessions straight from the left sidebar session menu (next to rename / fork / archive) and manage archived sessions from Settings -> Plugins -> Configurable. Head-frame only: no session log is ever decoded.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Github-CJX/dsh-session-dustbin@local/dsh-session-retry会话重试 · Session retry — 助手消息动作条上的「重新回答」按钮:先把这一轮的旧回答从模型上下文里移除,再把该轮的用户提问重新提交给同一个会话。 A DSH plugin that adds an "answer again" action to every assistant turn strip, dropping the previous answer from the model context first.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:linanbuan/dsh-session-retry@local/dsh-shell-settingsSettings → General row to pick the shell dsh runs agent commands with (system default / cmd / Windows PowerShell 5.1 / pwsh 7+).
@local/dsh-skill-managerDSH plugin: a Settings panel that lists the skills this Harness resolves, with each one's description, and pauses or resumes any of them by renaming the file DSH reads — no deletion, no restart
@local/dsh-skill-managerSettings page for viewing and managing existing skills: list all registered skills with source and invocation metadata, expand to read the full instructions, and delete user/project skills to the Windows Recycle Bin.
Use a feature keyword or category, then sort by recent updates or GitHub stars.
Read the description, AI analysis and original README to understand what a plugin does.
Review compatibility, dependency ranges and instructions, then copy the available install command.
DSH Hub is a community plugin directory for DeepSeek Harness, bringing together descriptions, source links and installation instructions to help you discover and compare extensions.
No. Detail pages show recorded compatibility and dependency ranges. Unknown does not mean compatible; check the project documentation before installing.
Open its detail page and check the installation section. Copy the command into your Harness environment when available, or follow the project README.