Token导航 LogoToken导航

Agent Skill · 研究检索

webfluence(webfluence 搜索)

webfluence 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。

webfluence 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,权限为权限未标注;许可证为 MIT,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 54,679。

总安装 499周安装 20GitHub Stars 175下载 162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
MIT

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:webfluence(webfluence 搜索)
来源仓库:https://github.com/nicepkg/ai-workflow
仓库路径:skills/webfluence
安装命令:
npx skills add https://github.com/nicepkg/ai-workflow --skill webfluence
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/nicepkg/ai-workflow --skill webfluence

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。

用于查找、检索和筛选相关信息。webfluence 属于研究检索类 Skill,可作为该场景下的辅助能力补充。适合根据关键词、任务场景或来源线索快速定位候选结果。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和维护状态,避免触发联网或文件读写。

涉及外部请求时应注意频率限制和数据隐私合规要求。

来源仓库内路径:skills/webfluence

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/nicepkg/ai-workflow --skill webfluence;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

    权限没有明确标注,第一次使用时只给一个无关紧要的任务,观察它到底读了哪些文件和网络。

常见问题

6 条

webfluence 怎么安装?

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

webfluence 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。

webfluence 需要什么权限?

权限级别没有明确标注。第一次使用时建议只给无关紧要的任务,观察它实际读写了哪些内容。

webfluence 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

webfluence 用的人多吗?

公开数据是累计安装 499 次,近一周安装 20 次,下载 162 次,GitHub 175 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。

webfluence 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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
webfluence

webfluence 搜索 · 本页

499175权限未标注MIT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

Webfluence

What This Does

Provides the Webfluence framework for architecting a content web that builds belief before asking for the sale. Use this to diagnose and fix content-to-conversion pathway problems.

The Core Problem This Solves

People send prospects directly to offer docs, VSLs, or sales pages without warming them up first. The offer doc is a closing tool, not a selling tool. Without sufficient "time on brand," conversion fails.

The Golden Rule

Nobody sees the offer doc without seeing content and a warming-up VSL before it, or they've come through a paid workshop or event.

The proper sequence is:

Content → VSL → Offer Doc
    OR
Workshop/Event → Offer Doc

Skipping steps = asking the offer doc to do work it wasn't built for.

The 47-Minute Rule

Research showed buyers consumed approximately 47 minutes of content before saying yes. Use this as a filter: "Is my web architected to deliver enough time on brand before asking for the sale?"

The Core Principle

When marketing builds belief, persuasion becomes unnecessary.

If your content does the heavy lifting, sales calls become easy (or unnecessary). Prospects arrive pre-sold, feeling like hot referrals.

The Five Layers of a Content Web

Layer 1: CORE (Center)

  • Purpose: Converts belief into buying intent
  • Example: VSL, paid workshop
  • Function: Your 24/7 salesperson that sells while you sleep

Layer 2: MIDDLE RING

  • Purpose: Builds authority and connection
  • Example: YouTube long-form, podcast, in-depth email sequences
  • Function: Deepens belief and establishes expertise

Layer 3: OUTER RING

  • Purpose: Drives reach and awareness
  • Example: Short-form content (Instagram Reels, LinkedIn posts)
  • Function: Entry points that create reach and recall

Layer 4: CONNECTORS

  • Purpose: Links everything together
  • Examples: ManyChat automations, GPT tools, email campaigns
  • Function: Qualifies and nurtures leads, moves traffic between layers

Layer 5: COMPOUNDER

  • Purpose: Keeps everything evergreen
  • Example: YouTube channel as long-tail belief machine
  • Function: When someone discovers one video, they binge the entire channel

Scoring a Content Web (Out of 5)

LayerQuestion
CoverageDo you have content for cold, warm, and hot audiences?
ConsistencyDo you show up rhythmically? (Not frequency, consistency)
ConversionDoes every piece of content have a clear CTA leading somewhere?
ConnectionDoes every piece physically link to another?
Belief FlowDoes your content build belief, not just awareness?

The two lowest-scoring areas are usually where the problem lives.

Common Diagnostic Patterns

Pattern: "Sent to offer doc, no sales"

Diagnosis: Skipping steps. Email → Offer doc bypasses content and VSL. Fix: Content → VSL → Offer doc. The email's job is to get them to content, not the offer doc.

Pattern: "High opens, low clicks, no sales"

Diagnosis: Missing connectors + possible list-offer mismatch. Fix: Check if CTAs drive to content that builds belief. Check if audience matches offer.

Pattern: "They watched the VSL but didn't apply"

Diagnosis: Insufficient belief-building before VSL, or VSL itself needs work. Fix: More content touchpoints before VSL. Review VSL for belief-building elements.

Pattern: "I don't have long-form content"

Response: You don't need all layers immediately. Start with what you have:

  • At minimum: Short-form + Email + Chat automation
  • YouTube/podcast can come later
  • Focus on connecting existing content before creating new

Implementation Priority

Don't try to build everything at once. Million Dollar Coach built this over 18+ months:

  1. Brand Width first - Repurpose existing content, extend indoctrination sequences
  2. Email consistency - Daily or regular rhythm (first cycles will be rough, that's normal)
  3. Short-form content - With clear CTAs to tools or longer content
  4. Paid workshop - Eliminates freebie chasers, builds belief
  5. VSL - Your 24/7 salesperson
  6. Long-form (YouTube) - The compounder that creates binge behavior

Key Mindset Shifts

  • From volume to connection: Goal isn't more content, it's connecting what you have
  • From linear to web: Accept that buyers won't follow your planned path
  • From attribution to attention: Stop obsessing over tracking; measure time on brand
  • From perfect to consistent: First attempts will suck; keep exercising the muscle

Related Trainings

Point members to these for deeper dives:

  • Brand Width - The foundational concept of time on brand
  • Salesless Experiment - Proof that marketing can replace sales calls
  • The System 2025 - The tactical ladder showing all the pieces
  • 80-20 Content Strategy - Content pillars and structure
  • The Social System - How to turn stories into content
  • VSL Training - How to create your video sales letter

When Answering Questions

  1. Identify where they are in the web architecture
  2. Diagnose which layer or connection is broken
  3. Prescribe the specific fix (not the whole system)
  4. Point to relevant training if they need to go deeper

Remember: The offer doc is a closing tool. If someone's sending cold traffic directly to it, that's the problem.

平台分布

OpenCode

27.23%
按下载量换算44

Claude Code

22.89%
按下载量换算37

Cursor

18.73%
按下载量换算30

Gemini CLI

13.68%
按下载量换算22

goose

7.2%
按下载量换算12

github-copilot

3.44%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

权限未标注

当前来源未能明确判断权限范围,默认进入异常复核队列。

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。

继续浏览同类 Skills