DSH Plugin HubDSH 插件市场

xueccci/dsh-prompt-lab

Plugin插件 Native原生 ⭐ 3 MIT Prompts & Skills提示词与技能

Agent-structured prompt optimizer for DeepSeek Harness (dsh). Modes: normal/slim/expand. Engines: local, hybrid, DSH, cloud API. Modal confirm with score and continue-optimize.

Project Overview项目介绍

Edited by Zhiwei Agent (zhiwei)由知微Agent(zhiwei)编辑

dsh-prompt-lab is an Agent-structured prompt optimizer plugin that exists specifically for the DeepSeek Harness (dsh) ecosystem, and it ships only as a DSH-native extension rather than as a standalone cross-platform product. It is installed through the DSH CLI with the command dsh plugin --profile web add github:xueccci/dsh-prompt-lab#v0.1.1, which fetches its bundle-manifest and loads it into the Web process. After restarting that profile and hard-reloading the browser with Ctrl+Shift+R, an 优化 (Optimize) button appears on the left of the composer; pressing Ctrl/Cmd+Enter triggers the same action.

The core flow begins when a draft is typed into the composer and the user clicks 优化. The plugin first rewrites the draft into an executable Agent pipeline covering role, task, context, constraints, and output format, then opens a confirm modal that shows the original, the result, five-layer diagnostic chips, and an output score. The left pane of the modal stays editable so the user can re-run 继续优化 on it, copy the result back via 回填输入, write the pane to the composer with 采用 (which only fires when the composer draft is unchanged and shows a 已写入 toast), copy the result to clipboard, or cancel/Esc to leave the original draft untouched. Three rewrite modes are offered: 普通优化 for full Agent structure, 精简 to extract the trunk (core task, hard constraints, deliverable), and 扩写 to expand only on needs the user explicitly named.

Four engines govern where the polishing work happens. 仅本地 runs offline rules only and never calls a model. hybrid, the default, builds a local skeleton first and only escalates to a configured provider plus model when both resolve. 仅DSH prefers the DSH model and degrades to local with a yellow banner if no model is set or the call fails. 云端 API talks to an OpenAI-compatible HTTP endpoint such as DeepSeek or a custom provider and likewise degrades to local with a yellow banner on failure. Idle timeout is 15 seconds and total timeout 60 seconds. Settings live under 设置 → 提示词工坊 and include engine, mode, cloud API base URL/key/model with a 测试连接 button, reasoning behavior (self-loop and shortcut toggles, shortcuts on by default), and an 关于 section. Cloud API keys stay in browser localStorage and are never written to workspace files; only local mode keeps the draft fully on-device, while hybrid, DSH, and cloud-API modes send the draft plus local skeleton to the configured provider. The project is MIT licensed, marked as an independent community plugin, and explicitly not affiliated with DeepSeek AI; development uses pnpm install, pnpm build (tsdown into lib/), and pnpm check for unit and syntax tests.

dsh-prompt-lab 是一款面向 DeepSeek Harness（dsh）的 Agent 结构化提示词优化插件，仅作为 DSH 原生扩展存在。它通过 dsh plugin --profile web add github:xueccci/dsh-prompt-lab#v0.1.1 安装，伴随 bundle-manifest 清单由 Web 进程加载，重启后在输入栏左侧出现「优化」按钮；快捷键为 Ctrl/Cmd+Enter。

核心功能围绕「优化」按钮与确认弹窗展开：先把草稿按 Agent 结构（角色／任务／上下文／约束／输出格式）改写，再以原稿｜结果｜五层诊断加输出分的形式呈现。弹窗支持继续优化、回填输入、采用（仅当输入框草稿未变时写回并 toast「已写入」）、复制以及取消/Esc（绝不改动草稿）。模式分普通优化、精简、扩写三种；引擎则提供仅本地、hybrid（默认）、仅 DSH、云端 API 四档，超时上限为 idle 15 秒、总 60 秒。

典型流程是输入草稿→点击优化→在弹窗中复核→选择继续优化／回填输入／采用／复制。设置页提供引擎、模式、云端 API（含测试连接）、推理行为（自循环、快捷键默认开启）以及关于信息。云端 API 密钥仅保存在浏览器 localStorage，不写入工作区文件。本地模式草稿不出本机，hybrid/DSH/云端 API 模式下草稿与本地骨架会发送给已配置的模型服务商；DSH 或云端失败时降级本地并显示黄条。项目为 MIT 许可的独立社区插件，与 DeepSeek AI 无从属关系，开发使用 pnpm install/build/check。

View on GitHub在 GitHub 查看

Recommended强烈推荐 Send to your DSH发给你的 DSH 安装 
请帮我安装这个 DSH 插件。安装前先完成【兼容性检查 + 安全性检查】，检查通过再动手。

插件：dsh-prompt-lab（xueccci/dsh-prompt-lab）
仓库：https://github.com/xueccci/dsh-prompt-lab
本站详情页：https://www.yhbd.top/plugins/xueccci-dsh-prompt-lab/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 MIT · ⭐ 3 · 最近提交 2026-09-22 · 主语言 JavaScript

按下面顺序执行，每步先把结论告诉我，再进入下一步：

【1 兼容性检查】
① 我这边：DSH 版本、Node 版本、操作系统、当前 profile（web / desktop）。
② 读它的 README、package.json、插件 manifest，列出它要求的 DSH 版本 / Node 版本 / 操作系统 / 外部依赖 / 需要另外先装的运行时。
③ 逐条比对，结论只写「满足 / 不满足 / 未知」三种；不满足的给出可行替代方案。
④ 检查是否和我已装的插件冲突：命令名重复、skill / tool 重名、端口占用、重复注册的 MCP server。

【2 安全性检查】
① 仓库可信度：和上面「本站登记」是否一致；star / fork 数、创建时间、最近提交，是否归档或长期停更。
② 安装脚本：逐行看 package.json 的 preinstall / install / postinstall，以及 install.sh、setup.ps1 之类脚本。出现 curl|bash、下载后直接执行、混淆代码、访问与插件功能无关的域名，立刻停下来告诉我，不要继续装。
③ 依赖：列出新增依赖，标出无人维护、或与知名包拼写近似的可疑包（typosquatting）。
④ 权限与副作用：它会读写哪些目录、访问哪些域名、需要哪些 DSH 权限（filesystem / network / shell / clipboard 等），以及怎么卸载和回滚。
⑤ 如果它要求 sudo / 管理员权限，或权限明显超出功能所需，先停下来问我。

【3 安装】
上面两步没有「不满足」和「高危项」时才执行；用官方推荐方式安装，不要自行提权。

【4 汇报】
用表格输出：检查项 / 结论 / 依据 / 是否需要我决策。拿不准的一律写「未知」并说明要我怎么确认——不要猜，也不要替我决定。

Send this message to DSH in your current session: it verifies compatibility and security first (answering met / not met / unknown item by item) and only installs once everything checks out — it will stop and ask you if it finds a high-risk item. The box scrolls; the copy is the full prompt. CLI install commands may not be accurate across systems, so DSH is the safer route.把上面这条消息直接发给当前会话里的 DSH：它会先核对兼容性与安全性（逐条给「满足 / 不满足 / 未知」），确认没问题再安装，有高危项会停下来问你。框内可滚动，复制到的是完整提示词；安装命令不一定准确，发给 DSH 更稳。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 warning1 项注意
  • Only 3 stars - very few users, little community feedback星标只有 3，几乎没人在用，遇到问题缺少社区反馈
DSH walks through these 9 checksDSH 会逐条核对这 9 项

Compatibility兼容性

  • DSH, Node, OS and profile requirementsDSH 版本 / Node 版本 / 操作系统 / profile 是否满足要求
  • External dependencies and runtimes (Electron / Python / Docker, ...)外部依赖与运行时（Electron / Python / Docker 等）是否齐备
  • Conflicts with installed plugins: command names, skill / tool names, ports, duplicate MCP registration与已装插件是否冲突：命令名、skill / tool 重名、端口占用、重复 MCP 注册

Security安全性

  • Repo matches the facts registered here; archived or abandoned?仓库是否与页面登记一致，是否归档或长期停更
  • Safety of preinstall / install / postinstall and install.sh / setup.ps1preinstall / install / postinstall 与 install.sh、setup.ps1 是否安全
  • curl|bash, download-then-execute, obfuscation, unrelated domains → stop immediatelycurl|bash、下载即执行、混淆代码、无关域名 → 立刻停止
  • Typosquatting or unmaintained packages among the new dependencies新增依赖里有没有 typosquatting 或无人维护的包
  • Requested permissions vs. what the feature actually needs申请了哪些权限、是否超出功能所需（filesystem / network / shell / clipboard）
  • Any sudo / admin requirement, plus uninstall and rollback是否要求 sudo / 管理员权限，以及卸载与回滚方式

Anything uncertain must be marked unknown with a note on how to confirm it. This site's signal screen is a static snapshot, not a security audit.拿不准的必须标「未知」并说明要我怎么确认。本站的信号筛查是静态快照，不能替代安全审计。

Or use CLI install (for developers)或使用命令行安装（适合开发者）

CLI Install命令行安装 

dsh plugin --profile web add github:xueccci/dsh-prompt-lab#v0.1.1

把 xueccci/dsh-prompt-lab 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

5.2 KB · 缓存于 2026-09-29 22:08:57 · 5.2 KB

dsh-prompt-lab

English | 中文

Agent-structured prompt optimizer for DeepSeek Harness (dsh).

Click 优化 in the composer to rewrite the current draft into an executable Agent pipeline — role / task / context / constraints / output format — then review the result in a confirm modal (original | result | five-layer chips + output score). Local rules always work offline; optional LLM polish uses the model already configured in DSH or a cloud API.

Data boundary: Local mode never leaves the machine. hybrid / DSH / cloud-API polish sends the draft and local skeleton to the configured provider.

What it does

Generic polish plugins dsh-prompt-lab
Method Ask the model to rewrite once Agent ladder structure first
Modes Many strategy labels 普通优化 / 精简 / 扩写
Offline Needs a model Local rules produce structured output
Confirm Silent write-back Modal + score + explicit actions
Failure Hard fail Degrade to local with a yellow banner

Modal actions

Action Behavior
继续优化 Re-run the same optimize pipeline on the editable draft pane
回填输入 Copy the result into the draft pane for secondary editing
采用 Write the draft pane back to the composer (已写入)
复制 Copy the result to clipboard
取消 / Esc Never touch the composer draft

Engines

Engine Behavior
仅本地 Offline rules, never calls a model
hybrid (default) Local skeleton → DSH polish when provider+model resolve
仅DSH Prefer DSH; on missing model / fail → local + yellow banner
云端 API OpenAI-compatible HTTP (DeepSeek or custom); fail → local + yellow banner

Timeouts: idle 15s / total 60s.

Install

dsh plugin --profile web add github:xueccci/dsh-prompt-lab#v0.1.1

Restart the profile’s Web process (and Ctrl+Shift+R in the browser), then verify:

dsh --profile web --dump-config | findstr dsh-prompt-lab

Usage

  1. Type a draft in the composer.
  2. Click 优化 (or Ctrl/Cmd+Enter).
  3. Review the modal; edit the draft pane, 继续优化, or 回填输入.
  4. 采用 writes the draft pane back only if the composer draft is unchanged.

Showing the opening section of the README — the full document lives in the repository以上为 README 开头摘要，完整文档在仓库内 · View the full README on GitHub →在 GitHub 查看完整 README →

渲染基于仓库 xueccci/dsh-prompt-lab 的 README.md。 在 GitHub 查看原文 →

Related相关推荐

← 上一个 Prev dsh-llm-ollama 下一个 Next dsh-goal-mode-enhance →