DeepSeek Harness Plugin Hub

发布与管理完整 Harness Profiles,发现适合你的插件。

探索

插件目录环境预设文档中心动态

社区

发布插件联系我们报告问题

相关链接

Plugin Hub GitHubDeepSeek Harness 官方项目系统状态隐私说明
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

独立、非官方社区项目,与 DeepSeek 官方无隶属、授权或背书关系。

Imagegen — DeepSeek Harness 插件(DSH Plugin)
← Plugins

@little-traincar/dsh-imagegen

Imagegen

用于 DeepSeek Harness 的图像生成工具——Doubao Seedream 5.0 Pro / qwen-image-3.0-pro,无水印,聊天内嵌显示。

插件会安装到这里;不确定时保持 web。

npx -y @deepseek-ai/dsh plugin --profile web add @little-traincar/dsh-imagegen@0.1.5
README兼容性版本

兼容性与来源证明

Imagegen 以 @little-traincar/dsh-imagegen 发布,当前版本为 0.1.5。Plugin Hub 会校验它的 manifest,并保存精确安装来源,便于复现安装结果。

DSH 兼容范围
*
运行环境
web
发布来源
npm
Registry 更新时间
2026/9/16

版本

0.1.5stable
2026/9/16
0.1.3stable
2026/9/9
0.1.2stable
2026/9/9

相关插件

正在加载相关插件…

最新版
0.1.5
DSH
*
HMR
重启进程
Tree shaking
未声明可安全裁剪
解包体积
51.8 kB
文件数
7
Surface
web
许可证
MIT
发布源
npm
GitHub
★ 1
周下载
0
最近提交
2026/9/16
查看源码 ↗
README Badge

点击下方 Badge 复制 Markdown,粘贴到 README 即可。

这是你的 Plugin?认领权益 · 优先安全扫描

验证 package.json 声明的 GitHub 仓库,即可管理这个公开页面。认领后,Hub 会优先安排当前版本的安全扫描,并在通过后公开展示结果。

认领这个 Plugin →
报告问题
DeepSeek Harness Plugin Hub
ProfilesPlugins分类动态文档登录管理 Profiles
ProfilesPlugins分类动态文档登录

README

English | 简体中文

dsh-imagegen---DSH RAW Image Plugin

Register the 'generate_image' tool for DeepSeek Harness agent: doubao Seedream 5.0 Pro / qwen-image-3.

setup

# npm 
dsh plugin --profile <name> add @little-traincar/dsh-imagegen

# Github
dsh plugin --profile <name> add github:little-traincar/dsh-imagegen#v0.1.1

# tarball
dsh plugin --profile <name> add ./dsh-imagegen-0.1.0.tgz

GitHub installation will be intercepted by pnpm for the first-time build authorization: add the package key to the profile's pnpm-workspace.yaml as prompted by dsh.

allowBuilds:
  '@little-traincar/dsh-imagegen': true

(Source code checkout users can also: pnpm dsh web -- patch<repository>/imagegen/codes. yml)

delete

# pnpm
dsh plugin --profile <name> remove @little-traincar/dsh-imagegen

use

1.Restart dsh ->Settings ->Imagegen ->Fill in Bean Pack/QWEN Key ->Save (with immediate effect; you can also use the environment variables' ARK_API_KEY '/' DASHSCOPE.API_KEY ').

2.In the conversation, it was mentioned that there is a need.

for example: Draw a summer coffee shop promotion poster: retro magazine collage style, cream yellow+coffee brown, vertical version; Text: Main title "Summer Ice Cafe Festival" sub title "Half price for the second cup of the entire event";

The image is embedded and displayed in the conversation, and saved to 'outdir' (default '/generated images', can be changed in settings or patches).

**Map generation * *: paste the map into the session (or give the local absolute path/URL) and say "change this map to China-Chic style/change the background" - the agent will redraw based on the reference map through the 'image' parameter.

Configuration (Priority: GUI Settings>cordis. catch. yml>Environment Variables)

itemdefaultexplanation
apiKeys.*NULLdoubao / qwen key;environment variable ARK_API_KEY / DASHSCOPE_API_KEY
baseUrls.*officialCan point to any OpenAI style /images/generations Compatible gateway
models.*snapshot IDDoubao alias will be 404, do not change it back doubao-seedream-5-0-pro
outDirUnder the startup directorygenerated-imagesDisk directory (absolute path)
count / aspect / attachToConversation / requestTimeoutMs2 / 2:3 / true / 300000picture habit

other

The dependent version is matched with the target DSH version; If the installation report says' version does not exist ', align the' package. json 'dependent version to your DSH version.

If you encounter the inability to input APIs, please check your DSH version. This plugin is developed based on 0.1.3-alpha. 2.

This project was completed with AI assisted development (Vibe coding) and has passed local smoke testing and real API call verification (real machine image output, watermark closure, attachment storage, and session embedding have all been tested and passed); If any abnormalities are found, please feel free to raise an issue or PR.

相关插件

继续浏览 vision-media 分类下经过校验的插件。

Tool Describe Image@linxin666/dsh-tool-describe-image面向模型的 describe_image 工具,用于 dsh Web GUI:通过在兼容 OpenAI 的端点调用视觉语言模型,为文本模型提供图像理解能力,以描述一张图像(本地路径、http(s) URL 或附件引用)。可热插拔 —Modlens@liustack/modlens面向仅支持文本的 LLM 的插件视觉能力,由免费的 Antigravity CLI 提供支持Deepseek Ivideodeepseek-ivideoiPolloWork HyperFrames Video Studio,以及 27 个可编辑视频模板,以原生 DeepSeek Harness 对话视图呈现。Image Gendsh-image-gen将类似 ChatGPT 的图像生成功能带到 DeepSeek Harness——支持 Gemini、OpenAI、Seedream、DashScope、本地 ComfyUI 等。