This skill should be used when the user wants to find leads, prospects, or contacts matching their ICP (Ideal Customer Profile). It searches for companies via Exa and enriches contacts via Apollo, outputting to CSV and optionally syncing to Attio CRM. MANDATORY TRIGGERS: "find leads", "prospecting", "ICP search", "find contacts", "lead generation", "/prospector"
数据来源:ClawHub。 在 ClawSkills 查看
选择你使用的 Agent
方法一:命令行安装(推荐)
推荐(无需提前安装 clawhub)
npx clawhub@latest --dir ~/.claude/skills install prospector或使用 clawhub CLI(需提前安装)
clawhub --dir ~/.claude/skills install prospector⚠️ 需要 Node.js 18+,没有 Node?请使用下方方法二直接下载 ZIP。 安装 Node.js →
方法二:手动下载安装(无需 Node)
下载 ZIP,解压后将文件夹放到以下路径,重启 Agent 即可:
安装路径
~/.claude/skills/prospector/💡解压后将文件夹放到上方路径,重启 Agent 即可生效
--- name: prospector description: | This skill should be used when the user wants to find leads, prospects, or contacts matching their ICP (Ideal Customer Profile). It searches for companies via Exa and enriches contacts via Apollo, outputting to CSV and optionally syncing to Attio CRM.
MANDATORY TRIGGERS: "find leads", "prospecting", "ICP search", "find contacts", "lead generation", "/prospector" version: 1.0.0 ---
Find leads matching your ICP via Exa company search + Apollo contact enrichment.
Run /prospector:setup first to configure your API keys:
You can also set keys via environment variables:
PROSPECTOR_EXA_API_KEYPROSPECTOR_APOLLO_API_KEYPROSPECTOR_ATTIO_API_KEY (optional)/prospector:setup
Collects and validates API keys, stores securely in ~/.config/prospector/config.json.
/prospector
Asks ICP questions, searches Exa, enriches via Apollo, outputs CSV to Desktop.
When the user invokes /prospector, follow this workflow:
First, verify env vars or config exist:
python3 -c "
import json
import os
from pathlib import Path
config_path = Path.home() / '.config' / 'prospector' / 'config.json'
env_exa = bool(os.getenv('PROSPECTOR_EXA_API_KEY'))
env_apollo = bool(os.getenv('PROSPECTOR_APOLLO_API_KEY'))
env_attio = bool(os.getenv('PROSPECTOR_ATTIO_API_KEY'))
if not config_path.exists():
print('NOT_FOUND')
else:
with open(config_path) as f:
config = json.load(f)
print('FOUND')
print(f'exa: {bool(config.get(\"exa_api_key\"))}')
print(f'apollo: {bool(config.get(\"apollo_api_key\"))}')
print(f'attio: {bool(config.get(\"attio_api_key\"))}')
print(f'env_exa: {env_exa}')
print(f'env_apollo: {env_apollo}')
print(f'env_attio: {env_attio}')
"
If NOT_FOUND and env vars are not set, tell user to run /prospector:setup first.
Use AskUserQuestion to collect ICP criteria in order:
Question 1: Industry
header: "Industry"
question: "What industry are you targeting?"
options:
- label: "SaaS"
description: "Software as a Service companies"
- label: "Fintech"
description: "Financial technology companies"
- label: "Healthcare"
description: "Healthcare and health tech"
- label: "E-commerce"
description: "Online retail and marketplaces"
- label: "AI/ML"
description: "Artificial intelligence and machine learning"
- label: "Any"
description: "No industry filter"
multiSelect: false
Question 2: Company Size
header: "Size"
question: "What company size are you targeting?"
options:
- label: "1-10"
description: "Early stage startups"
- label: "11-50"
description: "Seed to Series A"
- label: "51-200"
description: "Series A to B"
- label: "201-500"
description: "Growth stage"
- label: "500+"
description: "Enterprise"
- label: "Any"
description: "No size filter"
multiSelect: false
Question 3: Funding Stage
header: "Funding"
question: "What funding stage are you targeting?"
options:
- label: "Pre-seed"
description: "Pre-product market fit"
- label: "Seed"
description: "Building initial product"
- label: "Series A"
description: "Scaling product"
- label: "Series B+"
description: "Growth and expansion"
- label: "Any"
description: "No funding filter"
multiSelect: false
Question 4: Geography
header: "Geography"
question: "What geography are you targeting?"
options:
- label: "United States"
description: "US-based companies"
- label: "Europe"
description: "European companies"
- label: "Global"
description: "Worldwide"
- label: "Any"
description: "No geography filter"
multiSelect: false
Question 5: Keywords (optional)
header: "Keywords"
question: "Any specific keywords that should appear in company descriptions? (optional)"
options:
- label: "Skip"
description: "No keyword filter"
- label: "Enter keywords"
description: "I'll type specific keywords"
multiSelect: false
If "Enter keywords", ask for the text input.
Question 6: Contact Count
header: "Count"
question: "How many contacts do you want to find?"
options:
- label: "25"
description: "Quick search, lower API usage"
- label: "50"
description: "Balanced (recommended)"
- label: "100"
description: "Larger batch, more API usage"
multiSelect: false
Execute the Python script with the collected ICP:
cd [skill_directory]/scripts
python3 -c "
from prospector import run_search, export_csv, Config
icp = {
'industry': '[INDUSTRY]',
'company_size': '[SIZE]',
'funding_stage': '[FUNDING]',
'geography': '[GEOGRAPHY]',
'keywords': '[KEYWORDS or empty string]'
}
leads = run_search(icp, num_contacts=[COUNT])
if leads:
path = export_csv(leads)
print(f'SUCCESS: {len(leads)} leads saved to {path}')
else:
print('NO_RESULTS: No leads found matching criteria')
"
Replace placeholders with actual values from questions.
If Attio is configured and leads were found, ask:
header: "Attio"
question: "Sync leads to Attio CRM?"
options:
- label: "Yes"
description: "Sync companies and contacts to Attio"
- label: "No"
description: "Just keep the CSV"
multiSelect: false
If yes:
cd [skill_directory]/scripts
python3 -c "
from prospector import sync_to_attio, Config, Lead
import json
# Load leads from the CSV we just created (or pass them directly)
# For simplicity, re-run with sync
leads = [...] # Pass leads from previous step
companies, people = sync_to_attio(leads)
print(f'SYNCED: {companies} companies, {people} contacts')
"
Tell the user:
/prospector:setup安装 Prospector 后,可以对 AI 说这些话来触发它
Help me get started with Prospector
Explains what Prospector does, walks through the setup, and runs a quick demo based on your current project
Use Prospector to this skill should be used when the user wants to find leads, prospe...
Invokes Prospector with the right parameters and returns the result directly in the conversation
What can I do with Prospector in my marketing & growth workflow?
Lists the top use cases for Prospector, with example commands for each scenario
将技能文件夹放到 ~/.claude/skills/prospector/ 目录(个人级,所有项目可用),或 .claude/skills/prospector/(项目级)。重启 AI 客户端后,用 /prospector 主动调用,或让 AI 根据上下文自动发现并使用。
Prospector 支持 Claude、Cursor、OpenClaw,可与这些 AI 平台无缝集成,扩展其能力。
Prospector 可免费安装使用。请查阅仓库了解许可证信息。
This skill should be used when the user wants to find leads, prospects, or contacts matching their ICP (Ideal Customer Profile). It searches for companies via Exa and enriches contacts via Apollo, outputting to CSV and optionally syncing to Attio CRM. MANDATORY TRIGGERS: "find leads", "prospecting", "ICP search", "find contacts", "lead generation", "/prospector"
Prospector 属于「Marketing & Growth」分类,该分类的技能帮助 AI 智能体在此领域执行专业任务。
Automate my marketing & growth tasks using Prospector
Identifies repetitive steps in your workflow and sets up Prospector to handle them automatically