BillyChen123/qdd

Name: BillyChen123/qdd
Author: BillyChen123

Plugin插件 Native原生 ⭐ 7 Apache-2.0 Data & Analysis数据与分析 Prompts & Skills提示词与技能

Project Overview项目介绍

Edited by Zhiwei Agent (zhiwei)由知微Agent(zhiwei)编辑

This repository is a native DeepSeek Harness plugin that implements QDD, short for Question-Driven Discovery, a question-driven orchestration framework for AI-assisted biomedical research. It structures long-horizon research projects into an auditable loop of questions, evidence, artifacts, and next-study decisions. It ships with 34 role-routed domain skills covering common biomedical analysis tasks, and supports capturing public reference data from curated sources including CELLxGENE, GEO, PubMed, and CellMarker. It is currently at release candidate v0.1.0-rc.1, published as a stable reference for an accompanying academic paper, and includes a valid DSH bundle manifest.

QDD provides two modes of operation for research projects: guided human workflows and fully automatic mode. The six guided human workflows cover all stages of research from project initiation, study proposal, design exploration, task execution, study closure, all the way to final project conclusion. Automatic mode runs the core research loop continuously from project start through repeated study cycles, making it suitable for long-running automated research. It is designed specifically for biomedical researchers who want to keep their AI-assisted research organized, auditable, and fully reproducible.

QDD is released under the open-source Apache-2.0 license, and requires Node.js version 20.19.0 or newer to run. To install from source, users run three commands: npm install to pull dependencies, npm run build to compile the code, and npm install -g . to install a global qdd command line tool. Automatic mode requires an Anthropic-compatible model configuration, and DeepSeek models can be used via an Anthropic-compatible API gateway. QDD is explicitly not a clinical decision tool and does not replace domain expert judgment for research work.

QDD（Question-Driven Discovery）是面向AI辅助生物医学发现的问题驱动编排DSH原生插件，核心功能是将长周期研究转化为可审计的「问题-证据-研究产物-下一步决策」循环。它提供问题治理、人机可读代理内存、领域技能注入和公共数据接地四大核心能力，内置34个按角色和任务路由的本地领域技能，支持CELLxGENE、GEO、PubMed等公共生物医学数据的获取与归档。

QDD定义了6种标准化人机协作研究工作流，分别是项目启动、研究提案、方案探索、任务执行、研究收尾和项目总结，同时支持自动模式循环执行「启动-提案-执行-收尾」的研究循环，适合需要开展长周期多步骤研究的生物医学科研人员。它能帮助科研人员保持研究过程连贯可追溯，记录所有负向结果并支持基于结果的研究方向动态调整。

当前QDD处于v0.1.0-rc.1候选发布版本，采用Apache 2.0开源协议，运行要求Node.js版本不低于20.19.0，自动模式需要配置兼容Anthropic协议的大模型。它不是临床决策系统，也不能替代领域专家判断，安装需要通过npm构建并全局安装后，初始化项目即可使用自动模式或手动工作流。

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：qdd（BillyChen123/qdd）
仓库：https://github.com/BillyChen123/qdd
本站详情页：https://www.yhbd.top/plugins/billychen123-qdd/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 Apache-2.0 · ⭐ 7 · 最近提交 2026-08-19 · 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 7 stars - very few users, little community feedback星标只有 7，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 github:BillyChen123/qdd

把 BillyChen123/qdd 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

7.8 KB · 缓存于 2026-08-23 03:41:44 · 7.8 KB

Question-driven orchestration for AI-assisted biomedical discovery.

Turn long-horizon research into an auditable loop of questions, evidence, artifacts, and next-study decisions.

Language

简体中文 · English

Quick Links

Quick Start · Human Workflows · Auto Mode · Skills

Release status: QDD v0.1.0-rc.1 is the public submission-candidate line. This repository accepts reproducibility fixes, bug fixes, documentation improvements, and changes required by manuscript review. New research architectures are developed separately so this line remains a stable reference for the paper.

Quick Start

Requirements:

Node >=20.19.0
An Anthropic-compatible model configuration for Auto Mode

Install locally:

npm install
npm run build
npm install -g .

Initialize a research project:

mkdir my-qdd-project
cd my-qdd-project
qdd init .

Then either run the six human workflows through your agent, or start Auto Mode for the core research loop:

qdd auto --max-turns unlimited

More installation details are in docs/04-installation-guide.md.

What QDD Gives You

Question governance Every study records how the research question changed: refinement, confirmation, pivot, or dissolution.	Agent-ready memory Contracts, study files, task records, artifacts, and evolution history stay readable to both humans and agents.
Domain skill injection 34 local skills are routed by role and task instead of being dumped into every prompt.	Public-data grounding CELLxGENE, GEO, PubMed, CellMarker, and ligand-receptor references become auditable local artifacts.

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 BillyChen123/qdd 的 README.md。在 GitHub 查看原文 →