DeepSeek Harness plugin

live-talk

Animate any photo into a responsive virtual girl. She talks, turns, smiles, and moves naturally in sync with your conversation. Low-lag, high-detail.

Jump to install

Source facts

Repository
eddiehuang227-source/live-talk
Latest update
Aug 17, 2026
Category
Tools & Capabilities
GitHub stars
0
Format
plugin
Catalog evidence
Upstream dsh.bundle evidence
Evidence path
package.json#dsh.bundle
Checked against
0.1.0-rc.8
Upstream check date
2026-08-20

This evidence comes from the upstream catalog. This site has not installed, run, or security-reviewed the plugin.

Install

Start with a prompt that asks an agent to review the GitHub repository and source. Switch to the command if you want to install it yourself.

Copy this prompt into DSH, Codex, or another agent and ask it to review the GitHub repository and source first.

Do not install or run any commands yet. Read this plugin's GitHub repository, README, and relevant source code. Then answer the questions below clearly and directly so I can decide whether it fits my needs:

1. What is this plugin, and what problem does it solve?
2. Who is it for, and what are its typical use cases?
3. How is it used after installation? Include one minimal example.
4. What known limitations or privacy, security, compatibility, or maintenance risks does it have?
5. Give a clear recommendation: recommend, conditionally recommend, or do not recommend, with reasons.

Distinguish statements documented by the repository, inferences from source code, and unknowns. If evidence is insufficient, say so explicitly. Do not guess or simply repeat the README.

GitHub: https://github.com/eddiehuang227-source/live-talk
Plugin: live-talk
Author: eddiehuang227-source

Check the source files

Read the README and other files from this plugin directory before installing.

File explorer3 files
README.mdSource · read only

Live Talk

Animate any photo into a responsive virtual girl. She talks, turns, smiles, and moves naturally in sync with your conversation. Low-lag, high-detail.

三层结构

位置职责
dsh 插件壳(发布/安装/UI)package.jsoncordis.patch.ymlsrc/index.jssrc/client.jsdsh bundle/client manifest、host 路由、浏览器 conversation.view 标签页:选中角色舞台、开场视频、情绪 clip 联动与零 Key TTS 试听
对话角色可视化 coresrc/core/characters.jscharacter-registry.jsdialogue-pipeline.jsemotion.jsconversation-bridge.js人物 manifest、人物注册表、与供应商无关的语义管线(句子切分、情绪/动作、TTS 清洗)、dsh session 事件桥
语音/视频 provider seamsrc/core/seams.jsprovider-registry.jsasr / tts / avatar-media 三个可替换能力注册表

当前是能安装、能显示人物、语音/视频/实时 provider 齐备的切片:人物卡来自 /live/characters,UI 通过 dsh slot 注册;host 已注册 asr-doubao / tts-doubao / jimeng / realtime-volc / realtime-vidu provider(密钥走 ctx.credentials 或本机代理),client 提供零 Key browser-tts / browser-speech。剩余事项是 ctx.jobs 长任务接入与发布整理(见 [ROADMAP.md](ROADMAP.md))。

人物预览图与开场视频随包发布(assets/);情绪片段来自外部 character pack,缺失时 UI 自动回退到静态头像或占位图。素材授权边界见 [ASSET-LICENSE.md](ASSET-LICENSE.md)。

安装(当前 dsh 0.1.0-rc.5 验证)

dsh plugin --profile avatar add /path/to/live-talk
dsh --profile avatar --dump-config   # 应看到 live-talk 行
dsh --profile avatar web

打开 Web UI,在会话顶部视图标签选择 Live Talk

GitHub 直装需要 pin 提交并允许 prepare 构建脚本;npm/tarball 发布无需构建权限。

开发与测试

npm run build   # 从 src/ 生成 lib/(零依赖)
npm test        # build + 全部单元测试
npm run test:integration   # 需要 DSH_REPO 指向 deepseek-harness 源码

test:integration 会创建临时 DSH_HOME,安装 tarball,校验 --dump-config、启动 web 并请求 /live/health/live/characters 和客户端 bundle。

发布

  • 仓库添加 topic:dsh-plugin
  • npm:npm publishlib/ 已预构建);
  • tarball:npm pack,用户 dsh plugin add ./dsh-live-talk-0.6.0.tgz
  • 密钥类 provider 一律走 dsh ctx.credentials,不在 patch 或配置中携带;
  • 完整清单见 [PUBLISHING.md](PUBLISHING.md);素材边界见 [ASSET-LICENSE.md](ASSET-LICENSE.md)。

当前在 dsh 0.1.0-rc.5 上验证通过。