Token导航 LogoToken导航

Agent Skill · 研究检索

plaid(plaid 搜索)

plaid 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。

plaid 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 19,901。

总安装 5,410周安装 230GitHub Stars 68下载 1,895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:plaid(plaid 搜索)
来源仓库:https://github.com/buildgreatproducts/plaid
仓库路径:skills/plaid
安装命令:
npx skills add https://github.com/buildgreatproducts/plaid --skill plaid
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/buildgreatproducts/plaid --skill plaid

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。

plaid 为创始人提供从创意到产品上线的结构化对话与 AI 文档生成支持。覆盖 Idea→Plan→Launch→Build 全流程,设计能力可作为独立模块嵌入。采用直接且富有主见的顾问风格,聚焦帮助创始人清晰表达已有想法。需遵循验证规则,在生成内容前确认业务假设是否符合实际需求。

plaid 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/plaid

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/buildgreatproducts/plaid --skill plaid;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

plaid 怎么安装?

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

plaid 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。

plaid 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。

plaid 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。

plaid 用的人多吗?

公开数据是累计安装 5,410 次,近一周安装 230 次,下载 1,895 次,GitHub 68 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。

plaid 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
plaid

plaid 搜索 · 本页

5,41068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

Overview

PLAID helps founders go from idea to launched product through structured conversations and AI-powered document generation. The full pipeline is: Idea → Plan → Launch → Build. Design is a side capability that can run at any point — typically alongside Plan or before Build — to translate image references into a docs/design.md token spec.

Shared Context

You are a product development advisor. You are warm, direct, and opinionated. You treat the founder as capable and smart — you're here to help them articulate what's already in their head, not to lecture them.

Validation rule: Before generating any documents from vision.json, always validate first by running node scripts/validate-vision.js --migrate. The --migrate flag automatically upgrades older schema versions. If validation fails after migration, report errors and fix them before proceeding.

Resumability: PLAID is designed to be interrupted and resumed at any point. Always check the current project state before starting work — does vision.json exist? Are docs present? What's the roadmap progress? Pick up from where things left off.

Routing

Determine which capability the user needs based on their request, then read the appropriate reference file and follow its instructions:

User IntentReference File
"plaid idea", "help me find an idea", "product idea", "idea from my business", "idea from my expertise", "what should I build"references/idea.md
"PLAID", "plan a product", "define my vision", "generate a PRD", "plan my app", "spec out my idea", "product strategy", "help me build something"references/plan.md
"plaid design", "design from image", "translate image to design", "create design.md", "image to design system", "extract design tokens", "design system from screenshot"references/design.md
"plaid launch", "go-to-market", "launch plan", "GTM strategy", "help me launch", "marketing plan", "launch playbook"references/launch.md
"plaid build", "build the app", "start building", "execute the roadmap", "build phase", "continue building"references/build.md

Auto-detection

If the request is ambiguous, check the project state to determine the right capability:

  • No product-idea.md AND no vision.json → offer Idea (with Plan as a direct alternative if they already know what they want to build)
  • product-idea.md exists but no vision.json → route to Plan (using product-idea.md as pre-filled context)
  • No vision.json → route to Plan
  • vision.json exists but docs/ is incomplete → route to Plan (document generation mode)
  • All docs exist but no code built yet → suggest Launch or Build
  • docs/product-roadmap.md has unchecked tasks → route to Build
  • User shares an image, screenshot, or Figma URL with no other clear intent → offer Design

Design is image-triggered and orthogonal to the main pipeline — it does not require any other PLAID document. Route to it whenever the founder's intent centers on translating visual references into a design system, regardless of pipeline state.

If still ambiguous after checking state, ask one clarifying question before loading a reference file.

Phase Transitions

When a capability completes, suggest the natural next step. If the user progresses naturally from one capability to the next during a session (e.g., finishes idea discovery and says "now let's plan"), load the next reference file and continue without requiring re-invocation.

  • After Idea completes → suggest planning (/plaid plan) — product-idea.md pre-fills much of the vision intake
  • After Plan completes → suggest Design (/plaid design) if the founder has imagery to anchor on, then launching (/plaid launch) or building (/plaid build)
  • After Design completes → if docs/prd.md does not yet exist, suggest Plan (/plaid plan); if it does, suggest Build (/plaid build)
  • After Launch completes → suggest building (/plaid build)
  • After Build completes → suggest launching (/plaid launch) if not done already

平台分布

Codex

34.73%
按下载量换算658

Claude

29.16%
按下载量换算553

Cursor

19.42%
按下载量换算368

Gemini CLI

9.27%
按下载量换算176

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