Fantasality/dsh-origin-plugin 预览 preview

Fantasality/dsh-origin-plugin

DSH Origin插件:通过MCP从DeepSeek Harness AI聊天驱动Origin科学绘图。写入数据,绘制线图/散点图/柱状图,导出PNG/SVG格式。

Project Overview项目介绍

This is a DSH-native plugin that enables AI agents to automatically create scientific plots, run data analysis and deliver editable results via dialogue, leveraging a locally installed copy of OriginLab Origin. It supports dual access modes: inside DSH it runs as a native tool bridged by the official @deepseek-ai/dsh-mcp-client, and outside DSH it can run as a standard MCP server that works with any MCP-compatible client like Kimi Code, Cursor, and Claude Desktop. The plugin packs 62 tools, covering a wide range of 2D, 3D plot types, fine-grained editing, and common statistical analysis for scientific research.

It is targeted at researchers who need to publish high-quality scientific figures with Origin, and it supports two common workflows. For quick preview of data with clear semantics, users can get a finished plot with a single tool call. For publication-quality figures or multiple-group comparisons, the plugin uses a plan-confirmation workflow: it first generates a plotting plan with hashing to prevent stale data, asks for user confirmation if needed, then executes the plan, verifies the output, and finally exports the full delivery package. It also includes a built-in cookbook that provides recommended call chains and default parameters for eight common research scenarios.

To run the plugin, you need a Windows system with a licensed copy of OriginLab Origin (version 2018 or newer) and Python 3.10 or higher installed locally. The plugin creates its own isolated virtual environment during installation, so it will not modify or pollute your system Python environment. There are four installation methods to choose from: the recommended automatic installer detects existing AI tools and configures the plugin automatically, you can also use npx to launch the server directly, configure it manually, or install it as a DSH bundle with one click. It is released under the open-source MIT license, supports safe concurrent multi-session operations, and is free to use.

这是一个专为DeepSeek Harness(DSH)开发的原生插件,可通过AI对话驱动本地已安装的OriginLab Origin软件,实现自动科学绘图、数据分析与可编辑结果交付。它支持双模式接入,在DSH内通过官方MCP桥接作为原生工具使用,也可作为标准MCP服务,接入Kimi Code、Cursor、Claude Desktop等其他支持MCP协议的AI客户端。插件内置62个工具,覆盖从基础二维图、3D图到统计分析、细粒度改图等多种科研绘图需求。

面向需要制作期刊级科研图表的科研工作者,支持快速出图和正式发表两种工作流。对于数据语义明确的临时查看需求,可通过单次调用直接完成出图;对于正式发表级图表或多组对比需求,则需要先通过计划确认流程确认参数和需求,再执行绘图,最后导出包含可编辑工程文件的交付结果。插件内置场景速查工具,可快速获取不同场景下的调用链和推荐默认参数。

本插件仅支持Windows系统,需要用户预先安装正版OriginLab Origin(2018及以上版本均可正常运行),同时需要本地有Python 3.10及以上环境。插件自带独立虚拟环境,不会污染系统环境,用户可通过自动安装脚本一键配置到多个AI客户端,也支持手动配置或npx直接拉起服务。插件采用MIT许可证开源,完全免费使用,原生支持多会话并发安全操作。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 note1 项提示
  • 15 stars - an early-stage project星标 15,属于早期项目
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 add dsh-origin-plugin

把 Fantasality/dsh-origin-plugin 加入你的 DSH 配置(web profile)即可启用。

READMEREADME

DSH Origin Plugin · AI 对话驱动 OriginLab Origin 科学绘图

npm version npm downloads GitHub release GitHub downloads GitHub stars tools error codes platform Origin python license

下载与收录(国内直连,无需科学上网)

渠道 状态
🟢 npm 官方源 dsh-origin-plugin —— npm i -g dsh-origin-plugin(默认最新版)
🟢 npmmirror 淘宝镜像 镜像页 —— 国内加速,自动同步
🟢 GitHub Release Releases —— 源码 tgz + 说明
🟢 dshfind 目录源 已收录(is_plugin=true,市场内带安装命令)
🟢 DSH 1024Store 已收录(DSH 桌面端市场来源)
🟢 PyPI 包已构建于 dist/,token 到位即发布

适配:Windows 10/11 · Origin/OriginPro 2021+(2026 实测)· Python 3.10+ · 13 种 MCP 客户端 + 豆包/Cherry Studio/通义灵码等 UI 型客户端 · 纯本机 COM 通道, 数据不出本机、无外网依赖。

国内可用 · DSH 生态 · MCP 桥接通用 —— 让 AI 对话直接驱动本机 OriginLab Origin 自动画图、分析、交付。 纯本机 COM 通道:数据不出本机、无外网依赖、秒级响应(对比需要外网的远程 Origin 服务(如 EditaPlot 类),国内访问不稳定且数据要上传第三方;本插件用你已装的正版 Origin,没有网络与隐私问题)。

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 →

← 上一个 Prev deepseek-harness-android-app 下一个 Next dsh-improved-inline-edit →