메뉴
Lead enrichment pipelines: web scraping to structured prospect profiles, company intelligence signals, CRM population — Firecrawl, Clearbit, Apollo patterns
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/UitbreidenOS/UitKit --skill lead-enrichment명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SOC 직업 분류 기준
SKILL.md 표시 중
| name | lead-enrichment |
| description | Lead enrichment pipelines: web scraping to structured prospect profiles, company intelligence signals, CRM population — Firecrawl, Clearbit, Apollo patterns |
Raw Input (email / domain / LinkedIn URL)
↓
Step 1: IDENTIFY — resolve email to person + company
Step 2: ENRICH — fetch firmographic data (company, tech stack, headcount)
Step 3: SIGNAL CHECK — recent news, funding, hiring, exec changes
Step 4: SCORE — ICP fit score
Step 5: STORE — upsert to CRM with enriched fields
// Using Hunter.io API
async function resolveEmailToPerson(email: string): Promise<PersonData | null> {
const res = await (
)
data = res.()
(data.. !== )
{
email,
: data..,
: data..,
: data..,
: (email),
}
}
() {
res = (, {
: ,
: { : , : process..! },
: .({ email }),
})
res.()
}
import FirecrawlApp from '@mendable/firecrawl-js'
const firecrawl = new FirecrawlApp({ apiKey: process.env.FIRECRAWL_API_KEY })
async function enrichCompanyFromWebsite(domain: string): Promise<CompanyData> {
// Scrape the company website with structured extraction
const result = await firecrawl.scrapeUrl(`https://${domain}`, {
formats: ['extract'],
extract: {
schema: {
type: 'object',
properties: {
companyName: { type: 'string' },
description: { type: 'string' },
industry: { type: 'string' },
products: { type: 'array', items: { type: 'string' } },
techStack: { type: 'array', items: { type: 'string' } },
teamSize: { type: 'string' },
founded: { type: 'number' },
headquarters: { type: 'string' },
},
},
},
})
return result.extract as CompanyData
}
async function detectTriggerSignals(company: string): Promise<TriggerSignal[]> {
const signals: TriggerSignal[] = []
// Funding signals — check TechCrunch, Crunchbase
const fundingNews = await searchRecentNews(`${company} funding round 2026`)
if (fundingNews.length > 0) {
signals.push({ type: 'funding', relevance: 0.9, context: fundingNews[0].title })
}
// Hiring signals — engineering hires suggest growth/new projects
const hiringData = await checkLinkedInJobs(company)
if (hiringData.engineeringJobCount > 10) {
signals.push({ type: 'hiring_surge', relevance: 0.7, context: `${hiringData.engineeringJobCount} open engineering roles` })
}
// Tech stack changes — new tools = new needs
const techChanges = await checkBuiltWithHistory(getDomain(company))
if (techChanges.recentAdditions.length > 0) {
signals.push({ type: 'tech_adoption', relevance: 0.6, context: `Added: ${techChanges.recentAdditions.join(', ')}` })
}
return signals.sort((a, b) => b.relevance - a.relevance)
}
async function synthesiseProspectProfile(
person: PersonData,
company: CompanyData,
signals: TriggerSignal[]
): Promise<ProspectProfile> {
const { object } = await generateObject({
model: anthropic('claude-opus-4-7'),
schema: z.object({
icpScore: z.number().min(0).max(100),
painPoints: z.array(z.string()),
outreachAngle: z.string(),
bestChannel: z.enum(['email', 'linkedin', 'cold_call']),
bestTiming: z.string(),
notAGoodFit: z.boolean(),
notAGoodFitReason: z.string().optional(),
}),
prompt: `Analyse this prospect for a ${process.env.OUR_PRODUCT_DESCRIPTION}.
Person: ${person.firstName} ${person.lastName}, ${person.jobTitle} at ${company.companyName}
Company: ${company.description}. ${company.teamSize} employees. ${company.industry}.
Tech stack: ${company.techStack.join(', ')}
Recent signals: ${signals.map(s => s.context).join('; ')}
Score their ICP fit, identify pain points we can solve, and suggest the best outreach angle.`,
})
return { ...person, ...company, signals, ...object }
}
async function enrichLeadList(emails: string[]): Promise<EnrichedLead[]> {
const results: EnrichedLead[] = []
for (const email of emails) {
try {
console.log(`Enriching ${email}...`)
const [person, company] = await Promise.all([
resolveEmailToPerson(email),
enrichCompanyFromWebsite(getDomain(email)),
])
if (!person || !company) {
results.push({ email, status: 'not_found' })
continue
}
const signals = await detectTriggerSignals(company.companyName)
const profile = await synthesiseProspectProfile(person, company, signals)
// Upsert to HubSpot
await upsertHubSpotContact(email, profile)
results.push({ email, status: 'enriched', profile })
// Rate limiting
await new Promise(r => setTimeout(r, 500))
} catch (err) {
results.push({ email, status: 'error', error: err.message })
}
}
return results
}
User: Build a pipeline that takes 50 company domains from a CSV, scrapes each website for company data, detects funding and hiring signals, scores ICP fit, and pushes results to HubSpot.
Expected output:
scripts/enrich-pipeline.ts — reads domains.csv, runs enrichment, writes results.jsonenrichCompanyFromWebsite(domain) — Firecrawl structured extractiondetectTriggerSignals(company) — funding + hiring + tech signalsscoreICP(company, criteria) — 0-100 scoreupsertHubSpotContact(email, enrichedData) — creates/updates CRM recordsfailed-enrichments.csv