Token导航 LogoToken导航

Agent Skill · 研究检索

icp-matching(ICP 匹配)

icp-matching 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。

icp-matching 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 36,193。

总安装 1,968周安装 82GitHub Stars 85下载 656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:icp-matching(ICP 匹配)
来源仓库:https://github.com/guia-matthieu/clawfu-skills
仓库路径:skills/icp-matching
安装命令:
npx skills add https://github.com/guia-matthieu/clawfu-skills --skill icp-matching
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/guia-matthieu/clawfu-skills --skill icp-matching

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、检索和筛选与 ICP(目标客户画像)匹配相关的算法或规则。适合在 CRM 或推荐系统中实现用户与画像的自动对齐。需结合实际业务逻辑调整匹配权重和阈值参数。icp-matching 属于研究检索类 Skill,可作为该场景下的辅助能力补充。可通过 GitHub 仓库安装,集成至主流 AI 编程助手环境。

使用前应验证匹配结果的准确率,避免过度泛化或误判。

来源仓库内路径:skills/icp-matching

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/guia-matthieu/clawfu-skills --skill icp-matchi…;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

icp-matching 怎么安装?

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

icp-matching 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。

icp-matching 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。

icp-matching 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。

icp-matching 用的人多吗?

公开数据是累计安装 1,968 次,近一周安装 82 次,下载 656 次,GitHub 85 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。

icp-matching 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
icp-matching

ICP 匹配 · 本页

1,96885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

ICP Matching

Systematically evaluate prospects against your Ideal Customer Profile to prioritize high-fit accounts and improve SDR efficiency.

When to Use This Skill

  • Building lead scoring models
  • Qualifying inbound leads
  • Prioritizing outbound target accounts
  • Refining ICP definitions
  • Training SDRs on qualification

Methodology Foundation

Based on TOPO ICP Framework and Gartner B2B Buying Research, combining:

  • Firmographic fit (company attributes)
  • Technographic fit (tech stack)
  • Behavioral signals (intent data)
  • Persona fit (contact attributes)

What Claude Does vs What You Decide

Claude DoesYou Decide
Structures ICP frameworkICP criteria values
Scores prospects against ICPPass/fail thresholds
Identifies fit gapsPrioritization rules
Suggests disqualificationException handling
Ranks prospect listsTerritory assignments

Instructions

Step 1: Define ICP Dimensions

Firmographic Criteria:

AttributeIdealAcceptableDisqualify
Company Size100-100050-2000<25 or >5000
Revenue$10M-$500M$5M-$1B<$2M
IndustrySaaS, TechProfessional ServicesGovernment
GeographyUS, CanadaUK, EUAPAC
Growth StageSeries A-CSeed, Series D+Pre-seed

Technographic Criteria:

AttributeIdealAcceptableDisqualify
CRMSalesforceHubSpotNone
Tech StackModernMixedLegacy only
CompetitorsNone installedEvaluatingLocked in

Persona Criteria:

AttributeIdealAcceptableDisqualify
TitleVP/DirectorManagerCoordinator
DepartmentSales, MarketingRevOpsIT
Decision PowerBudget authorityInfluencerUser only

Step 2: Weight and Score

Scoring Model:

ICP Score =
  (Firmographic × 40%) +
  (Technographic × 30%) +
  (Persona × 30%)

Each dimension: 0-100 points

Score Interpretation:

ScoreFitAction
80-100ExcellentPrioritize immediately
60-79GoodStandard outreach
40-59FairNurture, low priority
<40PoorDisqualify

Step 3: Apply to Prospects

For each prospect:

  1. Gather available data
  2. Score each dimension
  3. Calculate weighted score
  4. Classify and prioritize

Examples

Example 1: ICP Definition

Output:

## Ideal Customer Profile: B2B Sales Tech Company

### Tier 1 (Excellent Fit - 80-100)
**These are your ideal targets. Prioritize always.**

**Company:**
- Size: 100-500 employees
- Revenue: $20M-$200M ARR
- Stage: Series B or C funded
- Industry: B2B SaaS
- Growth: 30%+ YoY
- Location: US/Canada

**Technology:**
- CRM: Salesforce
- Stack: Modern (cloud-native)
- Current solution: None or legacy competitor

**Buyer:**
- Title: VP Sales, CRO, Head of Sales
- Reports to: CEO/COO
- Budget: >$50K annually

---

### Tier 2 (Good Fit - 60-79)
**Worth pursuing with standard effort.**

**Company:**
- Size: 50-100 OR 500-1000 employees
- Revenue: $10M-$20M OR $200M-$500M
- Stage: Series A or D
- Industry: B2B Tech (not pure SaaS)
- Location: UK, EU

**Technology:**
- CRM: HubSpot
- Stack: Mix of modern/legacy
- Current solution: Considering change

**Buyer:**
- Title: Director of Sales Ops
- Reports to: VP Sales
- Budget: $25K-$50K annually

---

### Tier 3 (Fair Fit - 40-59)
**Nurture only. Don't prioritize.**

**Company:**
- Size: 25-50 OR 1000-2000 employees
- Revenue: $5M-$10M
- Stage: Seed
- Industry: Professional services

**Buyer:**
- Title: Sales Manager
- Budget: <$25K

---

### Disqualified (Score <40)
**Do not pursue. Auto-reject.**

- Company size <25 employees
- Revenue <$2M
- Industry: Government, Education, Non-profit
- Geography: APAC, LATAM (no support)
- Competitor locked in (multi-year contract)
- No CRM in use
- Title: Individual contributor

Example 2: Prospect Scoring

Input:

Score this prospect:

Company: GrowthTech Inc
Employees: 280
Revenue: $45M (estimated)
Industry: B2B SaaS
Founded: 2019
Location: Austin, TX
Recent news: Series B ($30M) in January

Tech Stack:
- CRM: Salesforce Enterprise
- Marketing: Marketo
- No sales engagement tool

Contact: Jennifer Wu
Title: VP of Revenue Operations
LinkedIn: Active, posts about sales productivity
Previous company: Used our competitor

Output:

## ICP Match: GrowthTech Inc

### Overall Score: 92/100 ⭐ EXCELLENT FIT

---

### Firmographic Score: 95/100

| Criteria | Finding | Points |
|----------|---------|--------|
| Size (280 emp) | ✅ Ideal (100-500) | 20/20 |
| Revenue ($45M) | ✅ Ideal ($20-200M) | 20/20 |
| Industry (B2B SaaS) | ✅ Ideal | 20/20 |
| Stage (Series B) | ✅ Ideal | 20/20 |
| Location (US) | ✅ Ideal | 15/15 |
| **Subtotal** | | **95/100** |

### Technographic Score: 90/100

| Criteria | Finding | Points |
|----------|---------|--------|
| CRM (Salesforce) | ✅ Ideal | 35/35 |
| Stack (Modern) | ✅ Marketo + SF | 25/25 |
| Competition | ✅ None installed | 30/30 |
| **Subtotal** | | **90/100** |

### Persona Score: 92/100

| Criteria | Finding | Points |
|----------|---------|--------|
| Title (VP RevOps) | ✅ Decision maker | 35/35 |
| Department | ✅ Revenue Operations | 25/25 |
| Activity | ✅ Active on LinkedIn | 15/20 |
| Previous | ✅ Knows competitor | 15/15 |
| **Subtotal** | | **92/100** |

---

### Weighted Total

(95 × 0.40) + (90 × 0.30) + (92 × 0.30) = 38 + 27 + 27.6 = 92.6 → 92/100

---

### Match Summary

**🎯 Tier 1: Excellent Fit**

**Strengths:**
1. Perfect company size/stage/industry
2. Salesforce user (easy integration story)
3. VP RevOps = ideal buyer persona
4. Recent funding = budget available
5. No competitor installed = greenfield

**Conversation Angles:**
1. "Congrats on Series B! How is that affecting your sales team scale?"
2. "I noticed you're not using a sales engagement tool yet—most RevOps leaders at your stage are building that out now."
3. "Given your Salesforce/Marketo stack, our integration would be seamless."

**Risk Factors:**
- None identified

**Priority: Immediate outreach**
Add to top of SDR queue. High-value, high-probability.

Skill Boundaries

What This Skill Does Well

  • Structuring ICP frameworks
  • Scoring prospects systematically
  • Identifying fit gaps
  • Prioritizing outreach

What This Skill Cannot Do

  • Access company databases
  • Verify data accuracy
  • Know internal buying dynamics
  • Predict conversion likelihood

When to Escalate to Human

  • ICP definition changes
  • Borderline accounts
  • Strategic target accounts
  • Exception requests

References

  • TOPO ICP Framework
  • Gartner B2B Buying Research
  • SalesLoft ICP Best Practices
  • Outreach.io Target Account Model

Related Skills

  • lead-scoring - RevOps scoring model
  • signal-monitoring - Intent data
  • prospecting-research - Deep account research

Skill Metadata

  • Domain: SDR Automation
  • Complexity: Intermediate
  • Mode: centaur
  • Time to Value: 2 hours for ICP, 5 min per prospect
  • Prerequisites: ICP criteria definition

平台分布

Codex

35.67%
按下载量换算234

Claude

33.72%
按下载量换算221

Cursor

17.79%
按下载量换算117

Gemini CLI

9.59%
按下载量换算63

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