Token导航 LogoToken导航

Agent Skill · 研究检索

discovery-research(发现研究)

discovery-research 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

discovery-research 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 61,604。

总安装 349周安装 15GitHub Stars 211下载 122核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:discovery-research(发现研究)
来源仓库:https://github.com/andreaskelm/pm-brain
仓库路径:skills/discovery-research
安装命令:
npx skills add https://github.com/andreaskelm/pm-brain --skill discovery-research
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/andreaskelm/pm-brain --skill discovery-research

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

discovery-research 作为产品探索阶段的智能协作者,支持用户需求调研与机会评估工作流。触发条件包括计划访谈、验证 idea 或分析流失原因等典型探索型任务。输出包含研究框架建议、问题清单与分析方法选择,侧重方法论而非具体答案。不替代人工调研,而是辅助构建系统性提问策略与资料整理规范。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/discovery-research

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/andreaskelm/pm-brain --skill discovery-researc…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

discovery-research 怎么安装?

在终端执行 npx skills add https://github.com/andreaskelm/pm-brain --skill discovery-research。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

discovery-research 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

discovery-research 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

discovery-research 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

discovery-research 用的人多吗?

公开数据是累计安装 349 次,近一周安装 15 次,下载 122 次,GitHub 211 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

discovery-research 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
discovery-research

发现研究 · 本页

349211只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Discovery & Research Skill

Use this skill when the user is exploring what to build or why something is happening, especially when they mention research, interviews, discovery, JTBD, opportunity assessment, validation, or product-market fit.

When to use this skill

Trigger this skill (in product_sense or execution_mode) when the user:

  • Says things like:

- “I need to talk to users” - “I need to plan research / interviews” - “I want to understand why users churn / don’t adopt” - “I want to validate this idea” - “I’m not sure if there’s real demand” - “We need to understand problem vs solution space” - “Are we at product-market fit yet?”

  • Or explicitly mentions:

- Interviews, discovery, JTBD, opportunity assessment, idea validation

If the user is mostly doing strategy (where to go), use the strategy-planning skill instead. If they’re mostly doing communication to stakeholders, use stakeholder-management.

Relevant framework locations

Most discovery frameworks live under:

  • 02-Methods-and-Tools/2.2-Discovery/

- 2.2.1-Research-Interviews/ - 2.2.2-Continuous-Discovery-Habits/ - 2.2.3-Jobs-To-Be-Done/ - 2.2.4-Opportunity-Assessment/ - 2.2.5-Idea-Validation/ - 2.2.6-Problem-Solution-Space/ - 2.2.7-Segmentation/ - 2.2.8-Product-Market-Fit/

Each follows the usual pattern:

  • 1-*-framework.md — guide + “For Agents” section
  • 2-*-template.md — template to fill out (when present)
  • 3-*-evaluation.md — optional evaluation guide (when present)

Typical flows

1. “I have an idea, should we build it?”

  1. Stay in product_sense first (golden rule: braindump before structure).
  2. Ask:

- What problem do you think this solves, and for whom? - What evidence or signals do you already have? - What’s at risk if you’re wrong?

  1. Then guide to:

- Problem/Solution separation: - 2.2-Discovery/2.2.6-Problem-Solution-Space/1-problem-solution-space-framework.md - Opportunity evaluation: - 2.2-Discovery/2.2.4-Opportunity-Assessment/1-opportunity-assessment-framework.md

  1. Only after sufficient braindump, move to execution_mode:

- Use the relevant 2-*-template.md to structure the opportunity.

2. “I need to run user interviews”

  1. Clarify:

- What do you want to learn or decide from these interviews? - Who exactly do you want to talk to? - What constraints do you have (time, access, volume)?

  1. Point to:

- 2.2-Discovery/2.2.1-Research-Interviews/1-interview-guide.md - 2.2-Discovery/2.2.2-Continuous-Discovery-Habits/README.md and the snapshot/synthesis steps.

  1. Help them:

- Define a focused research question. - Draft 6–10 core interview questions from the framework. - Plan what they’ll do with the data (snapshots → synthesis → opportunities).

3. “I want to understand product-market fit”

  1. Clarify product, segment, and current signals (retention, NPS, qualitative feedback).
  2. Point to:

- 2.2-Discovery/2.2.8-Product-Market-Fit/1-pmf-framework.md

  1. Decide together:

- Whether to start with qualitative (interviews) or quantitative (surveys, metrics), based on scale and access.

4. “I have JTBD job stories; how do I segment?”

  1. Clarify: Have they done 15–30+ job interviews? Do they need to decide who to serve first?
  2. Point to:

- 2.2-Discovery/2.2.7-Segmentation/README.md and 1-segmentation-framework.md

  1. Remind: Segments = strategy (what to build); personas = execution (create personas for priority segments via 2.3.5-Personas).

Response guidelines

  1. Think, then framework, then template

- In product_sense: stay with prompts and braindump before picking discovery frameworks. - Use prompts from 2-product-sense-prompts.md for discovery scenarios.

  1. Tie methods to constraints

- Always ask about time, access to users, and tooling before recommending heavy methods.

  1. Be specific

- Point to exact files: e.g., “From 02-Methods-and-Tools/2.2-Discovery/2.2.4-Opportunity-Assessment/1-opportunity-assessment-framework.md…”

  1. Connect the dots

- Cross-reference related frameworks (e.g., interviews → snapshots → opportunity assessment).

Relation to other skills

  • Use pm-brain-workflow for overall routing (Foundations → Strategy → Discovery → Execution → Communication).
  • Use this discovery-research skill for the discovery slice of that flow.
  • Use strategy-planning when the main question is direction and goals.
  • Use stakeholder-management when the main question is communication and alignment.

平台分布

Codex

35.58%
按下载量换算43

Claude

29.75%
按下载量换算36

Cursor

16.93%
按下载量换算21

Gemini CLI

8.83%
按下载量换算11

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills