Skills Plugins MCP Prompt Model 导航 博客 资讯 我的中心
界面增强 #canvas#deepseek-harness#dsh-plugin#particle-animation#typescript

dsh-whale-bg

非官方 DeepSeek Harness Web 界面粒子鲸鱼背景:光标照明、静止游动与随滚动渐隐。

gooosie @gooosie ⬇ 1 ★ 0 main

安装

dsh plugin --profile web add github:gooosie/dsh-whale-bg
下载安装清单

需要可复现安装时,可在仓库后追加 #commit 固定提交。

非官方 DeepSeek Harness Web 界面粒子鲸鱼背景:光标照明、静止游动与随滚动渐隐。

该插件未提供要点说明,请参考仓库 README。

canvasdeepseek-harnessdsh-pluginparticle-animationtypescript
  1. 安装并启动 DeepSeek Harness:npx @deepseek-ai/dsh web
  2. 在终端执行上面的安装命令(CLI 会解析插件并核验来源)
  3. 用 dsh plugins list 确认已安装,必要时重启 Harness 生效

插件以当前 dsh 进程的权限运行,安装时可能执行代码。请先通读仓库源码与许可证,确认无破坏性命令与越权访问;本站只做索引,不对第三方插件安全性作担保。

代码仓库github.com/gooosie/dsh-whale-bg
许可证MIT
主要语言main
下载量1
GitHub 星标0
最近推送2026-08-15
收录日期2026-09-19
分类界面增强

事实信息来自公开插件目录快照(2026-10-01),介绍文案由本站再加工。

以下为插件仓库 README 全文(原始内容,由公开目录抓取整理)。

# dsh-whale-bg

English | [简体中文](README.zh.md)

An unofficial DeepSeek Harness Web UI background that rebuilds the DeepSeek
fish silhouette as a swarm of varied-size particle tiles.

  [图片: Animated preview of the dsh-whale-bg particle effect]

> This is a community project. It is not affiliated with, maintained by, or
> endorsed by DeepSeek. The fish geometry is reused from the MIT-licensed
> DeepSeek Harness repository; see [THIRD_PARTY_NOTICES.md](THIRD_PARTY_NOTICES.md).

Features:

- **Entrance** — particles assemble from a scattered state into the whale
- **Idle swimming** — edges stay loose, the tail carries a travelling wave
- **Hover** — a radial mouse push with cubic falloff + per-tile angle noise
- **Flashlight lighting** — the light hugs the cursor: the whale is nearly
  invisible until the mouse comes near, then brightens inside a small radius
- **Themes** — light (multiply blend, deep blue) and dark (screen blend, glow)
- **30 FPS** rendering to keep the decorative layer inexpensive

## Compatibility

Developed and tested against `@deepseek-ai/dsh@0.1.0-rc.6` on its Web profile.
DeepSeek Harness is still in developer preview, so future releases may require
plugin changes.

## Install

Install the `dsh` CLI by following the
[DeepSeek Harness instructions](https://github.com/deepseek-ai/deepseek-harness#run).
Then install the published plugin into DSH's Web profile:

```sh
dsh plugin --profile web add dsh-whale-bg@0.1.0
dsh web
```

To install from a source checkout instead:

```sh
git clone https://github.com/gooosie/dsh-whale-bg.git
cd dsh-whale-bg
npm ci
npm run check
dsh plugin --profile web add .
dsh web
```

DSH's `web` command always boots the profile named `web`, so the plugin must be
installed into that profile. When installing from source, DSH links to the
checkout; do not move or delete that directory while the plugin is installed.

The `dsh.bundle` patch inserts the `whale-bg` row and the `dsh.client`
declaration mounts the browser half as `/plugins/dsh-whale-bg/client.js`.

Verify the composed configuration with `dsh web --dump-config`.

Remove:

```sh
dsh plugin --profile web remove dsh-whale-bg
```

## Development

```sh
npm ci
npm run build   # tsdown -> lib/index.js (ESM host) + wrapped lib/client.js
npm test        # validate manifests, entries, DSH handoff, injection, and theme API
npm run check   # strict types + clean build + package verification
npm run watch
```

## Structure

```
dsh-whale-bg/
├── package.json       # dsh.bundle (patch) + dsh.client (web module) manifests
├── cordis.patch.yml   # inserts the whale-bg composition row
├── tsdown.config.ts   # ESM host bundle + CJS client bundle
├── scripts/            # package-level regression verification
├── README.zh.md        # Simplified Chinese documentation
├── THIRD_PARTY_NOTICES.md
└── src/
    ├── index.ts       # no-op host entry (enables the Loader entry / scan)
    └── client.ts      # the whole whale background plugin
```

## Tuning

All knobs live in `src/client.ts`:

| Knob | Location | Default |
| --- | --- | --- |
| Particle density | `SAMPLE_SIZE`, `MODEL_STEP` | 60, 0.18 |
| Whale size / position | `computeLayout` | ≤720×540, center |
| Flashlight radius | `lightDistance / 7` | 7 model units |
| Idle visibility floor | `light = 0.05 + …` | 0.05 |
| Mouse push radius | `distance < 4.9` | 4.9 model units |
| Frame rate | `1000 / 30` | 30 FPS |

## Contributing and license

Contributions are welcome; see [CONTRIBUTING.md](CONTRIBUTING.md). Project code
is available under the [MIT License](LICENSE). Third-party notices and the
non-affiliation statement are in
[THIRD_PARTY_NOTICES.md](THIRD_PARTY_NOTICES.md).

数据来源:公开的 DeepSeek Harness 插件目录与各插件 GitHub 仓库。本站为独立第三方目录,与 DeepSeek、幻方(High-Flyer)及插件作者均无隶属或背书关系。

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。