Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问clear审计未展示

client-discovery客户发现

Agent Skill

client-discovery 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。

总安装

442

周安装

19

GitHub Stars

公开资料未说明

下载量

155
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

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:client-discovery(客户发现)
来源仓库:https://github.com/dvorkinguy/claude-skills-agents
仓库路径:skills/client-discovery
安装命令:
npx skills add dvorkinguy/claude-skills-agents --skill "client-discovery"
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。

AgentSkills.tonpx skills
npx skills add dvorkinguy/claude-skills-agents --skill "client-discovery"

简介

查找、检索和筛选相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中定位线索。
  • 使用 npx 命令添加技能,具体用法参考 README。
  • 安装命令:npx skills add dvorkinguy/claude-skills-agents --skill "client-discovery"。
  • 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。

SKILL.md

name
client-discovery
description
Sales qualification and client discovery. Use when starting a new client engagement, qualifying leads, identifying pain points, or calculating ROI for automation proposals. Generates discovery forms, qualification scores, and ROI estimates.

Client Discovery

Qualify leads faster and identify high-value automation opportunities.

Quick Actions

NeedAction
Discovery call prep→ Questions Below
Pain point mappingreferences/pain-points.md
ROI estimationreferences/roi-formulas.md
Full questionnairereferences/questions.md
Intake formassets/templates/discovery-form.md

Discovery Process

Phase 1: Pre-Call Research (5 min)

Before the call, gather:

  1. Company: Size, industry, tech stack (LinkedIn, Crunchbase)
  2. Contact: Role, tenure, decision authority
  3. Signals: Recent funding, hiring, complaints on G2/social
  4. Competition: Who else they might be talking to

Phase 2: Discovery Call Structure (30 min)

[5 min] Rapport & Agenda
[15 min] Pain Discovery
[5 min] Impact Quantification
[5 min] Next Steps

Phase 3: Core Discovery Questions

Business Context

  1. "Walk me through your typical [process] from start to finish."
  2. "Who's involved at each step?"
  3. "What tools do you use today?"

Pain Identification

  1. "What breaks most often?"
  2. "What takes longer than it should?"
  3. "What would happen if this didn't get done?"

Quantification

  1. "How many hours per week does this take?"
  2. "How many people are involved?"
  3. "What's the cost of an error here?"

Decision Process

  1. "Who else needs to approve this?"
  2. "What's your timeline for solving this?"
  3. "Have you tried to fix this before?"

Qualification Framework: BANT+

CriteriaQuestionGreenYellowRed
Budget"What would you pay to solve this?"Has budgetCan find budgetNo budget
Authority"Who approves this?"Decision makerInfluencerUser only
Need"How urgent is this?"Immediate painNice to haveExploring
Timeline"When do you need it?"This monthThis quarterSomeday
Pain ScoreHours × Cost × Frequency>$2k/mo waste$500-2k/mo<$500/mo

Scoring

Score each BANT+ criteria: Green=3, Yellow=2, Red=1

Total Score:
- 13-15: HOT - Prioritize, fast proposal
- 9-12: WARM - Nurture, build case
- 5-8: COLD - Qualify out or long-term

Pain Point Discovery

Universal Pain Signals

SignalWhat They SayReal Pain
Time waste"We spend hours on..."Manual data entry
Errors"Things fall through cracks"No system of record
Delays"It takes too long to..."Sequential bottleneck
Frustration"I hate doing..."Repetitive task
Risk"We worry about..."Compliance/quality

Probing Questions

When they mention a pain:

  1. "How often does this happen?"
  2. "What do you do when it happens?"
  3. "Who gets affected?"
  4. "What does it cost you?"
  5. "How long has this been going on?"

Pain Severity Matrix

           Low Frequency    High Frequency
High Cost   [AUTOMATE]       [URGENT]
Low Cost    [IGNORE]         [AUTOMATE]

ROI Quick Calculator

Input Variables

Hours/week on task: [H]
People involved: [P]
Hourly cost (fully loaded): [C]
Error rate: [E]%
Cost per error: [EC]

Formulas

Monthly Labor Cost = H × P × C × 4.33
Monthly Error Cost = (H × P × 4.33) × E × EC / 100
Total Monthly Waste = Labor + Error Cost

Automation Savings = Total × 80% (conservative)
Your Price Target = Savings × 15-25%
Payback Period = Setup Cost ÷ Monthly Savings

Example

Task: Manual invoice processing
Hours: 10/week
People: 2
Rate: $35/hr
Error: 5%
Error cost: $200

Labor: 10 × 2 × $35 × 4.33 = $3,031/mo
Errors: (10 × 2 × 4.33) × 5% × $200 / 100 = $87/mo
Total: $3,118/mo

Automation saves: $2,494/mo (80%)
Your price: $374-$624/mo (15-25%)

Vertical-Specific Discovery

E-commerce

  • "How do you handle out-of-stock notifications?"
  • "Walk me through an order from purchase to delivery."
  • "What happens when a customer disputes a charge?"

Professional Services

  • "How do you track billable hours?"
  • "What's your invoicing process?"
  • "How do new clients get onboarded?"

Logistics

  • "How do you notify customers of delays?"
  • "What happens when a delivery fails?"
  • "How do you optimize routes?"

Legal/Finance

  • "How do you track document versions?"
  • "What's your approval process?"
  • "How do you ensure compliance?"

See references/pain-points.md for complete vertical-specific questions.

Discovery Output

After call, document:

# Discovery Summary: [Company]

## Contact
- Name: [X]
- Role: [X]
- Authority: [Decision Maker / Influencer / User]

## BANT+ Score: [X/15]
- Budget: [G/Y/R] - Notes
- Authority: [G/Y/R] - Notes
- Need: [G/Y/R] - Notes
- Timeline: [G/Y/R] - Notes
- Pain: [G/Y/R] - $X/mo waste

## Top 3 Pain Points
1. [Pain] - [Impact] - [Urgency]
2. [Pain] - [Impact] - [Urgency]
3. [Pain] - [Impact] - [Urgency]

## Recommended Solution
[1-2 sentences on automation]

## ROI Estimate
- Current cost: $X/mo
- Automation saves: $Y/mo
- Price range: $Z/mo
- Payback: N days

## Next Steps
- [ ] Send proposal by [date]
- [ ] Follow-up call [date]
- [ ] [Other action]

## Notes
[Anything else important]

Red Flags (Disqualify)

  • No clear problem (just "exploring AI")
  • No budget and won't discuss
  • No authority and won't intro decision maker
  • Bad-mouthed previous vendors
  • Wants custom enterprise for SMB price
  • Endless "one more question"

References

TopicFile
Pain Points by Verticalreferences/pain-points.md
ROI Calculation Detailsreferences/roi-formulas.md
Full Question Bankreferences/questions.md

Templates

TemplateFile
Client Intake Formassets/templates/discovery-form.md

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

27.55%
按下载量换算43

Antigravity

21.24%
按下载量换算33

Gemini CLI

18.52%
按下载量换算29

windsurf

12.07%
按下载量换算19

OpenCode

7.06%
按下载量换算11

Codex

3.21%
按下载量换算5

安全审计

暂无安全审计结果可展示。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills