xyavid/dsh-prompt-polish

Name: xyavid/dsh-prompt-polish
Author: xyavid

Plugin插件 Native原生 ⭐ 2 MIT Prompts & Skills提示词与技能

Prompt polishing for the DeepSeek Harness web GUI — one click turns a rough composer draft into a clearer, more actionable prompt: explicit goal, concrete requirements, constraints, and acceptance criteria. The button sits right next to the model name, calls the model your session already uses, writes the result straight back into the composer, and then turns into an undo arrow.

catalog descriptioncatalog 简介 / catalog description：DeepSeek Harness Web 插件：输入框模型名旁的小星星按钮，一键把草稿改写成结构化的可执行提示词并写回

View on GitHub在 GitHub 查看

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-polish（xyavid/dsh-prompt-polish）
仓库：https://github.com/xyavid/dsh-prompt-polish
本站详情页：https://www.yhbd.top/plugins/xyavid-dsh-prompt-polish/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 MIT · ⭐ 2 · 最近提交 2026-09-28 · 主语言 TypeScript

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 2 stars - very few users, little community feedback星标只有 2，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 @xyavid/dsh-prompt-polish

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

READMEREADME

27.8 KB · 缓存于 2026-09-29 05:49:28 · 27.8 KB

dsh-prompt-polish

Prompt polishing for the DeepSeek Harness web GUI — one click turns a rough composer draft into a clearer, more actionable prompt: explicit goal, concrete requirements, constraints, and acceptance criteria. The button sits right next to the model name, calls the model your session already uses, writes the result straight back into the composer, and then turns into an undo arrow.

English | 简体中文

Why

A prompt that a coding agent can act on usually states what to build, where, under what constraints, and how success is judged. Most drafts don't — they are quick notes. This plugin adds a polish affordance to the dsh composer: it sends the draft through a low-temperature rewrite pass over the harness's own LLM service, fills the improved text back, and keeps the original one click away.

Nothing about your intent changes: the strategy is allowed to make the ask clearer, never to invent requirements. Your API key never reaches the browser, and no part of the flow touches the session's model history.

Features


✨ Composer button	A small sparkle button in the composer tool row, immediately left of the model name — always at hand, and it reflows with the row
🔁 Three states, one button	Idle (sparkle) → click to polish. Running (spinner) → click to cancel. Done (undo arrow) → click to restore the original draft. No extra panels, bars, or dialogs
🧠 Your session's model	The call goes through `ctx.llm` on the route your session already uses, falling back to the harness-wide default model (`agent-default-model`) — no second model to configure, no separate key
🔑 Zero credential setup	Calls ride the harness LLM service; keys come from the harness credential store and are used host-side only
🛡️ Draft safety	Empty, attachment-only, over-length, and command/reference-chip drafts are refused up front; a failed call never modifies what you typed
↩️ Two kinds of undo	Native `Ctrl+Z` (the write goes through the official editor API) and the button's own undo state
⚙️ Live settings	`enabled`, strength, temperature, token and length budgets, timeout, and the strategy prompt apply from Settings → 插件配置 with no restart
🌏 Language follows the draft	Chinese in → Chinese out, English in → English out; code, paths, identifiers, and error text are preserved verbatim
🚫 No fabrication	Expression-level rewriting only: no new requirements, no invented facts, no technology choices you did not imply
🩺 Runtime diagnostics	Client state is exposed on `window.__dshPromptPolish` (`stage` / `attempts` / `mounted` / `errors`) for one-command troubleshooting

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 xyavid/dsh-prompt-polish 的 README.md。在 GitHub 查看原文 →