Copy this instruction and paste it into your AI agent:
Install this skill: https://customermates.com/en/skills/raw/orthogonal-comprehensive-enrichment.mdComprehensive Enrichment — Person & Company
Maximum data + correctness. Use ALL relevant APIs, cross-reference results, flag conflicts.
1. Identifier Detection
Detect input type, then route:
| Input | Contains | Route |
|---|---|---|
@ | Person + Company (extract domain) | |
| LinkedIn person URL | linkedin.com/in/ | Person + Company (from results) |
| LinkedIn company URL | linkedin.com/company/ | Company only |
| Domain | *.com, *.io, etc. | Company only |
| Company name | No special pattern | Company only |
| Name + company | "John Doe at Stripe" | Person + Company |
| Twitter/X handle | @handle or x.com/ | Person + Company (from results) |
Person always cascades to company. Once person enrichment reveals their employer (company name, domain, or LinkedIn company URL), automatically run full company enrichment too. The only time you skip company is if you truly can't identify one. If LinkedIn person URL provided: use full URL for Fiber calls, extract username/slug for other endpoints.
2. Person Enrichment
Run ALL of these in parallel where possible. Collect everything, then compile.
2a. Full Profile & Contact Info
Fiber kitchen-sink (accepts LinkedIn URL, email, or name+company):
# By LinkedIn URL:
orth run fiber /v1/kitchen-sink/person --body '{"profileIdentifier": "https://linkedin.com/in/johndoe"}'
# By email:
orth run fiber /v1/kitchen-sink/person --body '{"emailAddress": "john@stripe.com"}'
# By name + company:
orth run fiber /v1/kitchen-sink/person --body '{
"personName": {"fullName": "John Doe"},
"companyName": {"name": "Stripe"},
"companyDomain": {"domain": "stripe.com"}
}'Nyne person search (async — deep work history, education, social):
# Step 1: POST to start search
orth run nyne /person/search -d '{"query": "John Doe Stripe"}'
# Step 2: Poll with GET using request_id
orth run nyne /person/search -q request_id=REQUEST_IDSixtyfour enrich-lead (AI-powered — slow, ~30-60s, but finds rich context):
orth run sixtyfour /enrich-lead --body '{
"lead_info": {"first_name": "John", "last_name": "Doe", "company": "Stripe", "linkedin_url": "https://linkedin.com/in/johndoe"},
"struct": {"work_email": "Work email", "personal_email": "Personal email (Gmail, etc.)", "phone": "Phone number", "title": "Job title", "bio": "Short bio"}
}'2b. Email — Find & Verify
Collect ALL emails — work AND personal. Many use cases (recruiting, etc.) need personal emails. Present each email with its type (work/personal) and verification status.
Find work email (cross-reference Hunter + Tomba):
# Hunter (returns work email)
orth run hunter /v2/email-finder --query domain=stripe.com first_name=John last_name=Doe
# Tomba (returns work email + sometimes personal)
orth run tomba /v1/email-finder --query domain=stripe.com company=Stripe first_name=John last_name=DoeFind personal email — these sources often return personal (Gmail, etc.):
# Tomba from LinkedIn (often returns personal email)
orth run tomba /v1/linkedin --query url=https://linkedin.com/in/johndoe
# Tomba enrich (returns all known emails for a person)
orth run tomba /v1/enrich --query email=john@stripe.comNyne person/search and Sixtyfour enrich-lead (Section 2a) also return personal emails — check their results.
Verify ALL found emails (run all three verifiers per email):
orth run hunter /v2/email-verifier --query email=john@stripe.com
orth run tomba /v1/email-verifier --query email=john@stripe.com
orth run fiber /v1/validate-email/single --body '{"email": "john@stripe.com"}'Verify every email found — work and personal. Run verifiers in parallel across all emails.
2c. Phone
orth run sixtyfour /find-phone --body '{
"lead": {"first_name": "John", "last_name": "Doe", "company": "Stripe"}
}'2d. Social Profiles & Activity
LinkedIn profile (Fiber):
orth run fiber /v1/linkedin-live-fetch/profile/single --body '{"identifier": "https://linkedin.com/in/johndoe"}'LinkedIn recent posts (Fiber):
orth run fiber /v1/linkedin-live-fetch/profile-posts --body '{"identifier": "https://linkedin.com/in/johndoe"}'Twitter/X activity (Nyne — async, returns tweets + engagement metrics):
# Step 1: POST with Twitter URL
orth run -X POST nyne /person/newsfeed -d '{"social_media_url": "https://x.com/HANDLE"}'
# Step 2: Poll with GET
orth run nyne /person/newsfeed -q request_id=REQUEST_ID2e. Open-Ended Research
orth run linkup /search --body '{
"q": "John Doe Stripe VP Engineering recent news interviews talks",
"depth": "deep",
"outputType": "sourcedAnswer"
}'2f. Compile Person Data
Cross-reference all API results. Merge name, title, emails (work + personal with verification status), phone, LinkedIn, Twitter, work history, education, and recent activity. When APIs disagree, keep both values with source labels. Once employer is identified, run full company enrichment (Section 3). See Section 5 for output formatting.
3. Company Enrichment
Run ALL of these in parallel where possible.
3a. Overview
Brand.dev (industry, size, description, logo):
# By domain (primary):
orth run brand-dev /v1/brand/retrieve --query domain=stripe.com
# By company name (if no domain):
orth run brand-dev /v1/brand/retrieve-by-name --query name=Stripe
# By email (extracts domain):
orth run brand-dev /v1/brand/retrieve-by-email --query email=john@stripe.comHunter company data:
orth run hunter /v2/domain-search --query domain=stripe.comFiber company data (LinkedIn-enriched):
orth run fiber /v1/kitchen-sink/company --body '{"companyDomain": {"domain": "stripe.com"}}'3b. Leadership & Employees
Key people by title:
orth run fiber /v1/natural-language-search/profiles --body '{"query": "CEO, CTO, CFO, COO, VP at Stripe", "pageSize": 10}'3c. Funding
Nyne funding history (async):
# Step 1: POST
orth run -X POST nyne /company/funding -d '{"company_name": "Stripe"}'
# Step 2: Poll with GET
orth run nyne /company/funding -q request_id=REQUEST_IDNyne investors:
orth run -X POST nyne /company/funders -d '{"company_domain": "stripe.com"}'3d. Products & Web Presence
Products from website:
orth run brand-dev /v1/brand/ai/products --body '{"domain": "stripe.com"}'Scrape for pricing/features:
orth run scrapegraph /v1/smartscraper --body '{
"website_url": "https://stripe.com/pricing",
"user_prompt": "Extract all products, pricing tiers, and features"
}'Find competitors/similar companies:
orth run exa /findSimilar --body '{
"url": "https://stripe.com",
"numResults": 10,
"contents": {"text": true}
}'3e. Open-Ended Research
orth run linkup /search --body '{
"q": "Stripe recent news funding announcements partnerships press releases",
"depth": "deep",
"outputType": "sourcedAnswer"
}'3f. Compile Company Data
Cross-reference all API results. Merge overview, leadership, funding, products, competitors, news, and social presence. When APIs disagree, keep both values with source labels. See Section 5 for output formatting.
4. Full Pipeline Example — enrich john@stripe.com
Step 1: Detect — Email → person enrichment + extract domain stripe.com for company.
Step 2: Person enrichment (run in parallel):
# Profile (3 sources)
orth run fiber /v1/kitchen-sink/person --body '{"emailAddress": "john@stripe.com", "companyDomain": {"domain": "stripe.com"}}'
orth run nyne /person/search -d '{"query": "john stripe.com"}'
orth run sixtyfour /enrich-lead --body '{"lead_info": {"email": "john@stripe.com", "company": "Stripe"}, "struct": {"work_email": "Work email", "personal_email": "Personal email", "phone": "Phone", "title": "Title", "bio": "Bio"}}'
# Find personal email
orth run tomba /v1/enrich --query email=john@stripe.com
# Verify work email (3 sources)
orth run hunter /v2/email-verifier --query email=john@stripe.com
orth run tomba /v1/email-verifier --query email=john@stripe.com
orth run fiber /v1/validate-email/single --body '{"email": "john@stripe.com"}'
# Also verify any personal emails found with the same 3 verifiers
# Phone
orth run sixtyfour /find-phone --body '{"lead": {"email": "john@stripe.com", "company": "Stripe"}}'
# Research
orth run linkup /search --body '{"q": "john stripe.com", "depth": "deep", "outputType": "sourcedAnswer"}'Once you have the person's full name + LinkedIn from Step 2, fire off:
# LinkedIn profile + posts
orth run fiber /v1/linkedin-live-fetch/profile/single --body '{"identifier": "LINKEDIN_URL"}'
orth run fiber /v1/linkedin-live-fetch/profile-posts --body '{"identifier": "LINKEDIN_URL"}'
# Twitter (if discovered)
orth run -X POST nyne /person/newsfeed -d '{"social_media_url": "https://x.com/TWITTER_HANDLE"}'Step 3: Company enrichment (run in parallel with person):
# Overview
orth run brand-dev /v1/brand/retrieve --query domain=stripe.com
orth run hunter /v2/domain-search --query domain=stripe.com
orth run fiber /v1/kitchen-sink/company --body '{"companyDomain": {"domain": "stripe.com"}}'
# Leadership
orth run fiber /v1/natural-language-search/profiles --body '{"query": "CEO, CTO, CFO, COO, VP at Stripe", "pageSize": 10}'
# Funding
orth run -X POST nyne /company/funding -d '{"company_name": "Stripe"}'
orth run -X POST nyne /company/funders -d '{"company_domain": "stripe.com"}'
# Products & competitors
orth run brand-dev /v1/brand/ai/products --body '{"domain": "stripe.com"}'
orth run scrapegraph /v1/smartscraper --body '{"website_url": "https://stripe.com/pricing", "user_prompt": "Extract all products, pricing tiers, and features"}'
orth run exa /findSimilar --body '{"url": "https://stripe.com", "numResults": 10}'
# News
orth run linkup /search --body '{"q": "Stripe recent news funding announcements", "depth": "deep", "outputType": "sourcedAnswer"}'Step 4: Compile & Format — Merge all results, cross-reference, flag conflicts, then present using the two-tier output format (Section 5): summary card first, full details below.
5. Output Format
Always present results in two tiers: a scannable summary card on top, then full details below.
Tier 1: Summary Card
Lead with this. A sales rep should be able to scan it in 30 seconds.
For a Person (+ their company):
## 🔍 {Full Name} — {Title} at {Company}
**Contact**
- ✉️ Work: {email} ({verification status})
- ✉️ Personal: {email} ({verification status})
- 📱 {phone}
- 🔗 LinkedIn: {url}
- 𝕏 Twitter: {url}
**Bio**: {One-liner from best available source}
**Personalization Angles**
1. {Recent activity, talk, post, or news mention — with date}
2. {Another angle}
3. {Another angle}
**Company Snapshot**: {Company} · {industry} · {employee count} employees · HQ: {location}
Latest funding: {round type} — ${amount} ({date}) · Total raised: ${total} · Valuation: ${valuation}For a Company (standalone):
## 🏢 {Company Name}
**Overview**
- 🌐 {domain}
- 🏷️ {industry}
- 👥 {employee count} employees
- 📍 HQ: {location}
**Funding**: {latest round} — ${amount} ({date}) · Total raised: ${total} · Valuation: ${valuation}
**Key Decision Makers**
| Name | Title | Email |
|------|-------|-------|
| {name} | {title} | {email} |
| ... | ... | ... |
**Recent News & Icebreakers**
1. {headline — date — source}
2. {headline — date — source}
3. {headline — date — source}Tier 2: Full Details
Below a clear separator (---), include the complete deep-dive for those who want to dig in:
- Person: Full work history, education, all social profiles, all LinkedIn posts, all tweets + engagement, publications, full Linkup research, all email sources + verification breakdown
- Company: Full description, all funding rounds with dates/amounts/investors, complete leadership list, products + pricing tiers, competitor analysis, full news results, social presence stats
Present Tier 2 with clear section headers. Include source labels on every data point. Flag all conflicts between APIs.
6. Tips
- Parallelize: Run all independent API calls concurrently — person and company enrichment can run simultaneously
- Nyne is async: POST returns
request_id, poll with GET until status is complete (5-20 seconds) - Conflicts: When APIs disagree, show both values with source labels — never silently pick one
- LinkedIn URLs: Dramatically improve match rates for Fiber and Tomba — extract from any source that returns them
- All emails matter: Always collect both work AND personal emails — recruiting and hiring use cases need personal emails. Label each as work/personal
- Email verification: Verify every email (work + personal) with all 3 verifiers (Hunter, Tomba, Fiber) and take consensus
- Person → Company: Person enrichment always cascades — once you identify their employer, run full company enrichment automatically
- Linkup deep search: Best for personalization angles — recent talks, interviews, blog posts, news mentions
- Sixtyfour enrich-lead is slow: Takes 30-60 seconds (AI web research). Fire it early, don't block on it — continue processing other API results and merge Sixtyfour data when it arrives
- Nyne newsfeed for Twitter/X: Pass the Twitter URL to get recent tweets + engagement. Async like other Nyne endpoints
- Adaptive: Skip APIs that don't apply (e.g., don't run email-finder if email is already known, don't run funding search for public megacorps)
- Tomba linkedin: If you have a LinkedIn URL but no email, Tomba's LinkedIn finder is very effective
- Company from email: Brand.dev's retrieve-by-email endpoint handles domain extraction automatically