dsh-prompt-studio：DSH 提示词多风格优化与发送前自动改写
dsh-prompt-studio 是 DeepSeek Harness 的提示词优化插件（v1.0.0，维护者 zywnb-2）。在对话输入框提供「优化」面板，按内置或自定义风格改写草稿，支持撤销；开启「发送前自动优化」后，模型收到改写内容而会话仍保留原话。设置页可管理风格库、优化模型与温度等，配置存于 $DSH_HOME/prompt-studio/config.json。适用于 Windows 10/11 与 DSH 桌面版 0.2.0-rc.2，推荐从 GitHub 安装 github:zywnb-2/dsh-prompt-studio#v1.0.0 并重启 DSH。
Read Moredsh-prompt-polisher：DSH 聊天框一键优化提示词
dsh-prompt-polisher 是 yuluo554 维护的 DeepSeek Harness web 插件，在 Composer 输入框增加「优化提示词」按钮，支持会话模型与 OpenAI 兼容独立 API 双通路、concise/structured 风格与竞态保护。本文介绍安装命令 dsh plugin --profile web add dsh-prompt-polisher、配置项与使用限制（引用 chips、最短长度、dsh 0.2.x）。
Read Moredsh-input-optimizer：发送前重写草稿，并管理对话分支
dsh-input-optimizer 是 Xxsttb 维护的 DeepSeek Harness 社区插件（模型推理）。在 Composer 模型选择器左侧提供四角星按钮，用当前对话模型以低推理强度重写未发送草稿，修正语法、歧义与步骤顺序，且不生成回答、不进会话日志。另支持已发用户消息分叉编辑、无记忆重新生成、右侧记忆树及在工具改写工作区时禁用铅笔与重生成。要求 DSH ≥0.2.0-rc.2，通过 dsh plugin add 本地路径或 pnpm pack 的 tgz 安装，设置页可配重写指令、模型路由与编辑快捷键。
Read Moredsh-input-optimizer：在输入框里用 Harness 已有路由改写提示词
dsh-input-optimizer 是 xkjxziochu 维护的独立 DSH 插件，在输入框发送按钮旁提供提示词优化：通过 ctx.llm 复用 Harness 已配置的 LLM 路由改写草稿，确认面板展示原文与优化后对照，采纳或取消。可配置 optimize 路由、Reasoning effort、自定义 system prompt 及 0–20 轮上下文。推荐安装：dsh plugin --profile <profile> add dsh-input-optimizer。需 dsh ^0.2.0-rc.1，自 dsh-better-input 拆出，不含语音识别。MIT 0.1.0。
Read Moredsh-composer-prompt-optimizer：在 DSH 输入框一键优化提示词
subisle 维护的 DSH 插件 dsh-composer-prompt-optimizer（GitHub：dsh-prompt-optimizer）在 Composer 工具栏提供一键提示词优化：结合最近用户轮次与会话上下文，复用当前选中模型发起单次 compaction 辅助调用，将草稿整理为目标、范围、要求、交付与验收等结构，不写入会话日志、不激活 Agent。支持 npm / tarball / 本地 / github 安装，可选 maxDraftChars、maxContextTurns 等配置。适用于常写 coding-agent 任务说明的 DSH 用户；安装前请审阅源码与 MIT 许可证。
Read Moredsh-prompt-wisp：在 DSH 输入框旁一键把草稿优化成结构化提示词
dsh-prompt-wisp（提示词精灵）是 jr-create 发布的 DSH 客户端插件：在 composer 发送键旁提供 ✦ 按钮，用宿主已配置模型将草稿改写为结构化提示词，支持简便/详细双模式与会话上下文感知；另在输入区上方提供死锁、无效输出、工具循环等只读监控告警。零运行时依赖，安装后重启 dsh web 即可使用。本文基于官方 README 介绍安装命令、设置项、API 与兼容性（DSH 0.1.5-rc.2、web profile）。
Read Moredsh-prompt-optimizer：在 DSH 输入框里用当前模型优化草稿
dsh-prompt-optimizer 是 EmotionG 维护的 DSH 常驻 bundle 插件，在 conversation.input.right 提供普通模式与优化模式切换。优化模式下基于当前会话上下文与会话所选模型，通过 /prompt-optimizer RPC 流式优化 composer 草稿并 setDraft 回填，支持 Alt+O 与可传播到底层传输的取消链路。本文依据 GitHub README 说明与动态插件版的差异、上下文与截断限制，以及 web profile 下 file: 依赖、dsh.profile.bundles 与 dsh plugin add 的安装步骤。
Read Moredsh-prompt-optimizer：输入框一键优化提示词草稿
kee0012 维护的 DSH 客户端插件 dsh-prompt-optimizer，在对话输入框工具栏（模型选择左侧）提供 ✨ 一键优化：用当前会话所选模型改写草稿并直接写回，原文备份可还原。风格（精简/标准/详细）、上下文、预览确认等收在 ▾ 弹出层并跨会话记忆。安装：dsh plugin --profile desktop add github:kee0012/dsh-prompt-optimizer，需重启 DSH。服务端 POST /dsh-prompt-optimizer/optimize，temperature 0.4，MIT 0.3.1。
Read Moredsh-prompt-lab：在 DSH 输入栏把草稿改写成 Agent 结构化提示词
dsh-prompt-lab（提示词工坊）是 xueccci 维护的 DeepSeek Harness 社区插件，在 Web composer 提供「优化」入口，将草稿改写成 role/task/context/constraints/format 结构。支持普通优化、精简、扩写三种模式，以及仅本地、hybrid、仅 DSH、云端 API 四种引擎；确认弹窗含五层诊断与输出分，支持继续优化与显式写回。需 DSH >= 0.2.0-rc.2，安装命令为 dsh plugin --profile web add github:xueccci/dsh-prompt-lab#v0.2.0。MIT 许可，与 DeepSeek AI 无隶属关系。
Read MoreAI漫剧道具一致性提示词优化 Skill：把分镜提示词对齐到指定年代
SkillHub 社区 Skill @user_32b1aab1/bai-fengxi（v1.0.1，design-media）用于优化 AIGC 漫剧提示词，按 SKILL.md 流程解析目标年代、抽取道具、校验时间线并替换穿越物品，输出优化提示词与修改说明表。无需 API Key。安装：skillhub install bai-fengxi --namespace user_32b1aab1 --dir 对应 Agent 的 skills 目录；亦可按 skillhub.cn/install/skillhub.md 让 Agent 安装。适合分镜、场景与道具清单写作，史实与生成效果仍建议人工复核。
Read More提示词优化工具：把模糊需求写成结构化 AI 提示词
SkillHub 社区 Skill「提示词优化工具」（@user_7871dce1/prompt-optimizer-zh，v1.0.0）将模糊需求转为精确、结构化的 AI 提示词，基于 EARS（易用需求规格语法），适用于写作、代码生成与产品功能描述等 prompt 场景；标签 requires_api_key=false。本文说明定位、能力与通过 skillhub CLI 安装到 Cursor/Claude Code/Codex 等 skills 目录的步骤，并给出与官方简介一致的使用方式与注意事项。
Read MoreDeepSeek Harness 提示词润色插件 (dsh-plugin-prompt-polish)
DeepSeek Harness 插件 dsh-plugin-prompt-polish 由开发者 imlida 维护，旨在解决调试智能体或提示词时反复手动改写的痛点。该插件在输入框右下角增加「✨润色」按钮，利用模型能力一键改写草稿，支持撤销、自定义润色指令及零污染对话历史。安装需通过命令行添加依赖并重启。配置灵活，支持修改提示词及更换模型。已知限制包括文件引用丢失、质量依赖模型以及暂不支持自动更新。
Read Moredsh-prompt-forge: DSH Desktop Prompt Optimization Plugin
The DeepSeek Harness desktop plugin dsh-prompt-forge enables one-click prompt optimization through an inline panel. After users enter a draft, the plugin reuses the host’s model channel and provides side-by-side comparison, streaming output, and continued refinement, supporting result adoption or one-click copying. Designed to be non-invasive, the plugin does not store keys and offers extensive customization options, making it an effective tool for improving prompt-writing efficiency.
Read MoreDeepSeek Harness Prompt Optimization Plugin (dsh-prompt-optimization-master)
The DeepSeek Harness (DSH) prompt optimization plugin is inspired by the Tencent WorkBuddy mechanism and aims to address the issue of unclear drafts. The plugin invokes a model independently to rewrite drafts and supports one-click restoration. It features intelligent state management (such as switching between idle and busy states), automatic reset, and session isolation, while optimizing output budget handling for reasoning-oriented models. The plugin supports both Web and desktop platforms, offers flexible configuration, and includes a comprehensive test suite covering host and client logic, as well as reasoning budget testing.
Read MoreDSH Prompt Optimizer
The DSH plugin `dsh-prompt-optimizer` is a client-side tool that injects a `♂harden` button into the bottom-left corner of the input box, enabling one-click rewriting of drafts into well-structured, assertive prompts. It supports three styles—tough, team, and cool—and features zero dependencies, local fallback, and HTTP interfaces. Installation requires Node >= 18; desktop users must manually refresh the page in Settings. Configuration options include style, maximum input characters, and maximum concurrency. The plugin provides `/health` and `/optimize` endpoints and strictly adheres to factual fidelity.
Read MoreDeepSeek Harness (DSH) Prompt Optimizer Plugin
The DeepSeek Harness plugin `co1ombiagly/prompt_optimize_dsh` aims to address the high barrier to entry in prompt engineering. It can convert casual requirements into rigorously structured prompts with one click and automatically insert them into the input field. Featuring context awareness, iterative rewriting, and zero API key management, it ensures data security. Users can quickly generate high-quality prompts with the `/optimize` command, without needing to manage an API Key.
Read MoreDeepSeek Harness Plugin: narrative-prompt-polish
`narrative-prompt-polish` is a prompt polishing plugin for DeepSeek Harness (DSH), designed to address insufficient clarity in drafts. Users can click the `✨` in the bottom-right corner of the main box to invoke a conversational sub-agent for multi-round follow-up optimization, then manually copy and paste the result back once satisfied. The plugin supports Prompt Cache–optimized computation and includes built-in intent classification and safe write-back mechanisms to prevent accidental operations. The repository has now been migrated to `dsh-prompt-only-forge` and depends on the DSH kernel’s sub-agent orchestration service.
Read Moredsh-prompt-boost-pro: Enhance DSH Prompts Before Sending
`dsh-prompt-boost-pro` is an MIT-licensed plugin integrated into DeepSeek Harness (DSH), designed to optimize prompts through visual comparison. It offers two modes—Structured (restructuring task objectives) and Light (refining wording)—and supports Chinese and English interfaces along with a smart undo mechanism, ensuring users can preview enhancements before confirming. Users can use it by installing the plugin and restarting the service, making it suitable for scenarios requiring precise control over instruction output.
Read Moredsh-prompt-refine: DeepSeek Harness Prompt Patch Optimization Plugin
dsh-prompt-refine is a DeepSeek Harness (DSH) plugin designed to automatically patch missing elements in user prompts through context-aware refinement. The plugin provides a ✨ button next to the send button in the input box. When clicked, it invokes the configured model to analyze the current draft and conversation history, generating three suggestions that include identified issues and suggested patches. Users can select the desired suggestions, preview the merged prompt in real time, and insert it into the input box, enabling lightweight, non-intrusive prompt optimization without additional configuration.
Read Moreprompt-polish: DeepSeek Harness Prompt Optimization Assistant
The DSH ecosystem plugin prompt-polish is designed to address vague input drafts using a prompt engineering agent. It supports one-click polishing and completion, multi-round iterative rewriting, and version rollback, with features including automatic input-box population, built-in diagnostic logging, and a zero-build workflow. Users can install it via the dsh command, configure an OpenAI-compatible API (such as DeepSeek or Ollama), or use the built-in routing, making it convenient to use and easy to uninstall.
Read MoreDSH-Prompt-Optimization: One-Click Prompt Optimization Plugin for DSH Web
DeepSeek Harness (DSH) has launched the Prompt-Optimization plugin, designed to address issues caused by imprecise prompt writing and provide native prompt optimization within the input box. The plugin supports one-click silent optimization, undo actions, and model configuration, and is easy to install for DSH Web users. When using it, attention should be paid to edge cases such as reference loss, plain-text enhancement, and request limits, including an input limit of 256KB.
Read MoreDeepSeek Harness Plugin: dsh-prompt-optimizer
The `dsh-prompt-optimizer` plugin, developed by Y1X1n, is designed to address the time-consuming and inconsistent manual prompt polishing in DeepSeek Harness. It adds an “Optimize” entry in the send bar and uses model routing to reuse the session model for assisted rewriting. Core features include a five-dimensional diagnostic view, dual-strategy optimization (structuring / refinement and polishing), fidelity discipline self-checking, context-aware memory, and safe session isolation. Installation requires configuring the dsh CLI first. Users can optimize with one click via the button, and the plugin supports configurable parameters and slash-command protection to ensure the output remains semantically equivalent without breaking command structure.
Read Moredsh-prompt-optimizer: One-Click Prompt Optimization in the DSH Input Box
DeepSeek Harness (DSH) has released the `dsh-prompt-optimizer` plugin to address the cumbersome process of writing prompts. By adding an icon button to the input field toolbar in the DSH Web interface, the plugin uses the currently active model to optimize user input with one click. It supports streaming backfill, preserves the original task intent and language style, and provides loading animations and empty-content protection. Users simply type a prompt into the input field and click the button to get clearer, more specific optimized results. The plugin is released under the MIT license, can be installed via the CLI, and is ready to use after restarting the Web interface.
Read Moredsh-prompt-optimizer: DSH Web GUI Input Prompt Optimization Plugin
`dsh-prompt-optimizer` is a plugin for DeepSeek Harness (DSH) designed to address the challenge of repeatedly refining prompts. It provides an “✦ Optimize” button, supports manual and pre-submission automatic optimization, and includes a comparison and rollback feature between the original text and the optimized result. The plugin is compatible with DSH’s built-in routing and OpenAI-compatible interfaces, securely stores API keys, and offers configuration options. It is easy to install, has low performance overhead, and is licensed under the MIT License.
Read Moredsh-prompt-boost: An Input Optimization Tool for the DSH Web Interface
`dsh-prompt-boost` is a plugin for DeepSeek Harness (DSH) designed to improve the quality of prompt inputs. It offers one-click rewriting, turning rough drafts into clear, structured instructions, and supports undo and language locking. Installation requires manual YAML configuration, the plugin accepts up to 8,000 characters of input, and it can be used through the browser web interface.
Read MoreDSH Prompt Polisher: A Native Prompt Polishing Plugin for DeepSeek Harness
`dsh-prompt-polisher` is a native prompt polishing plugin for DeepSeek Harness. By adding a "Polish" button to the input field, it reuses the current session configuration to optimize prompts directly, eliminating the need to switch environments. The plugin supports conflict detection, session isolation, result preview, and one-click application, while providing privacy protection. Users only need to install the required version and adjust parameters (such as context length and timeout) via the configuration file to debug prompts safely and efficiently in the DSH interface.
Read Moredsh-prompt-optimizer: One-Click Prompt Optimization and Long-Text Input Enhancement
The DeepSeek Harness (DSH) Web plugin aims to improve long-text input and prompt-structure issues. Core features include: a new “Optimize” button in the input box, which can use an LLM to rewrite prompts; support for configuring model routing and thinking level; and a fix for the Composer component’s rendering defect with long text. Additionally, the plugin supports indentation with the Tab key and increases the maximum height to `min(55vh, 600px)`. The plugin is configured via `cordis.patch.yml` and is ready to use after installation, but the optimization feature is disabled when U+FFFC attachment placeholders are present.
Read Moredsh-assistant-optimization: DSH web plugin for collapsible model output and enhanced prompts
# Introduction After starting a service using the web profile of DSH (DeepSeek Harness), several issues are encountered almost daily: - Some OpenAI-compatible APIs lack a dedicated `reasoning_content` channel; the Chain of Thought is mixed directly into the content stream, and DSH renders the entire block as plain text, burying the answer within a large chunk of thinking. - Once a turn is complete, the screen is cluttered with lines of thinking, tool calls, and intermediate explanations; what is often truly relevant is only the final reply. - If the process is manually stopped or the session encounters an error, resuming requires manually reconstructing the input. - When the model outputs Mermaid charts, the interface displays only a block of code text. dsh-assistant-optimiz
Read Moredsh-better-input: Add voice input, AI polish, and file-to-Markdown for DeepSeek Harness
dsh-better-input is an input enhancement plugin for the DeepSeek Harness (DSH) Web UI, maintained by DIAG5 and licensed under the MIT License. The plugin aims to optimize the input paths for long prompts, local documents, and dictated content. Its main features include: 1. **Voice Input:** Local recognition based on the Web Speech API, requiring no API Key, featuring real-time streaming transcription, and supporting custom recording durations. 2. **AI Polishing and Optimization:** Reuses the models configured in DSH to automatically clean up speech recognition errors, remove filler words, and precisely optimize prompts, supporting original text comparison and undo operations. 3. **Prompt Templates:** Templates are stored locally; typing `/` allows for quick insertion, ensuring data privacy. 4. **File Handling**
Read Moredsh-prompt-polish: DeepSeek Harness Input Bar Prompt Optimization Plugin
dsh-prompt-polish is a web-based prompt optimization plugin for DeepSeek Harness (DSH). Integrated with Composer, it allows users to click the Optimize button to send the current draft to a configured model service, generating a clearer and more specific version. The plugin supports six rewrite modes (e.g., default, concise, structured), Chinese/English language selection, and optional context injection such as objectives, to-do items, and conversation summaries. Optimization results are displayed in a popup with a comparison of the original and optimized drafts; users must explicitly click "Adopt" to apply the changes, which only takes effect if the draft remains unchanged. The plugin features runtime limits (such as draft length and timeout mechanisms), streaming stop, a history of the last five attempts, and local settings synchronization. The UI is designed to match the DSH theme and supports keyboard shortcuts and accessibility settings.
Read Moredsh-smart-input: Voice Input and Prompt Optimization Plugin for DeepSeek Harness
dsh-smart-input is a DeepSeek Harness (DSH) Web client plugin maintained by v-quest123456 under the MIT license, designed to address the issues of incomplete manual prompt writing and the lack of feedback in voice-to-text. Its core features include voice input and prompt optimization. Voice input supports the browser Web Speech API, local Whisper ONOX, and the online Whisper API, featuring real-time streaming transcription, smart punctuation, multi-language support for Chinese, English, and Japanese, and a continuous conversation mode, with the shortcut key being Ctrl/Cmd + Shift + V. Prompt optimization offers six strategies, including automatic, simplification, and structuring, supporting context-awareness (injecting workspace file information), before-and-after comparison, and one-click acceptance, with the shortcut key being Ctrl/Cm.
Read MoreDSH-AI-Prompt-Optimizer: Optimizing Prompts on the DSH Web Chat Page
The dsh-ai-prompt-optimizer is a prompt optimization plugin specifically designed for the DeepSeek Harness (DSH) web chat page, aiming to help users organize rough draft text into well-structured, complete AI prompts. Maintained by wuchubuzai2018, this plugin is licensed under the MIT License. Its core functionality is to read the raw text from the input field, invoke the default model configured in DSH to generate an optimized version, and then fill the result back into the input field for user confirmation before sending, rather than automatically submitting it. Key advantages include no need for a separate API Key configuration, reuse of existing model settings, visualized process status, and clear error notifications. Additionally, it improves compatibility with various model providers by sending only essential request fields. Installation requires Node.js 22+ and p
Read Moreoss-prompt-optimizer: Optimizing Raw Instructions in DSH into Structured Prompts
`oss-prompt-optimizer` is a prompt optimization tool within the DSH plugin ecosystem, designed to rewrite brief instructions into professional, structured prompts to address output generalization issues. The plugin operates through harness's LLM service without direct API connections or credential exposure, and it is licensed under MIT. Core functionalities include: providing the `prompt_optimize` tool for agents to support single-time optimization and iterative rewriting; offering a `ctx.promptOptimizer` programming interface for other plugins; integrating a one-click optimization feature in the composer input box; enabling automatic optimization triggered by the `/optimize` prefix (enabled by default, effective only for specific prefixes); and possessing contextual and situational awareness capabilities to automatically inject roles, tasks, and other elements.
Read Moredsh-prompt-optimize: Prompt Optimization, Personalized Instructions, and Usage Cards for DSH Web
`dsh-prompt-optimize` is a DSH web plugin designed to address the issues of scattered functionalities in draft rewriting, long-term preference settings, long conversation navigation, and API usage viewing. The plugin offers four core features: firstly, prompt optimization, which utilizes the DeepSeek model to rewrite drafts with a preview-before-apply option; secondly, personalized custom instructions, allowing users to set and persistently save long-term preferences; thirdly, user question anchors, enabling quick navigation and scrolling to specific user messages in long conversations via anchors on the right side or top; and fourthly, a DeepSeek API floating usage card, which displays real-time spending and balance for the current key over today, yesterday, and the last seven days, with data calculated in GMT+8. The plugin is licensed under MIT, requires the DSH environment to operate, and supports installation and uninstallation via the `dsh plugin` command.
Read More