Find People
/find-peopleSearch people across many companies at once and get back matching profiles with name, title, company, location and LinkedIn URL.
Overview
find_people is the SyncGTM MCP tool that searches professional profiles across many companies at once and returns name, title, company, location and LinkedIn URL for every match. Filters stack across role, seniority, company domain, headcount, location, tenure, education, skills and timing signals, so a broad ICP narrows into a list a rep can actually work. It returns profiles only, no emails and no phones, so you spend enrichment credits on the shortlist instead of the whole search.
What it does
- 1
Takes filters, not an identifier. No parameter is strictly required, but a call with no filters has nothing to search on — pass at least one.
- 2
Stacks filters across role (current_titles, seniority_levels, current_functions), company (current_company_domains, industries, current_company_types), size (current_company_headcount_ranges), location (person_countries, person_locations), tenure, education and skills.
- 3
Reads timing signals: recently_changed_jobs finds people new in seat, has_verified_business_email keeps only profiles that already have a verified business email on record, open_to accepts CAREER_INTEREST, HIRING_MANAGER or VOLUNTEERING.
- 4
Returns name, title, company, location and linkedin_url for each profile, plus a next_cursor to fetch the following page.
- 5
Does not return emails or phone numbers. Chain find_work_email and verify_email once you have picked the profiles worth contacting.
- 6
Costs 0.3 credits per profile returned. limit accepts 1–100 and defaults to 25, so a default call is 7.5 credits and a full page of 100 is 30.
How to use it
- 1
Download the skill file — it saves as find-people.zip.
- 2
In Claude, open Settings → Capabilities → Skills and upload the .zip (or unzip the find-people/ folder into .claude/skills/ for Claude Code).
- 3
Connect the SyncGTM MCP server at docs.syncgtm.com/mcp_server/setup — browser sign-in, no API key. Run check_credits (free) to confirm the connection works.
- 4
Type /find-people and describe the person in plain English. The skill maps the description onto real parameters — seniority_levels, current_company_headcount_ranges, person_countries — and reads the mapping back to you before it spends anything.
- 5
Set limit deliberately. It is the cost dial at 0.3 credits per row, so test a new filter combination at limit: 25 and only then page for volume.
Use cases
Build an ICP list
Turn a written ICP into a named list of prospects with LinkedIn URLs. Replaces a static database export you have to re-buy every quarter.
New-in-seat outreach
Set recently_changed_jobs to true to catch buyers in their first months, when tooling and budget decisions reopen. Pair with a seniority filter so you only get decision makers.
Map a buying committee
Pass several domains to current_company_domains and pull directors and VPs across all of those accounts in one call, instead of researching each company by hand.
Track champions who moved
Use past_company_domains with your customer domains to find former users now sitting at new accounts. The warmest cold list a rep can build.
GTM workflow examples
Monday ICP list build to verified outbound
- 1Run `check_credits` (free) to confirm the balance covers the run before anything bills.
- 2Run `find_people` with current_titles ["VP Sales", "Head of Sales"], industries ["Software Development"], current_company_headcount_ranges ["51-200"], person_countries ["United States"], has_verified_business_email true, limit 50 — returns 50 profiles with name, title, company, location and LinkedIn URL for 15 credits.
- 3Cut to the 25 best-fit rows on title and company, then run `find_work_email` on each LinkedIn URL — 1 credit each, produces business addresses.
- 4Run `verify_email` on every address returned — 0.3 credits each, removes invalid and undeliverable rows before they hit a sequence.
Outcome: A CSV of up to 25 rows — name, title, company, LinkedIn URL and a verified work email — built for at most 47.5 credits in one sitting.
Account-first displacement play
- 1Run `find_companies` with your firmographic filters and limit 40 to produce 40 target domains with employee count range and country, for 12 credits.
- 2Run `find_company_techstack` on those domains to see which ones already run the competitor you displace, and drop the rest.
- 3Run `find_people` with current_company_domains set to the qualifying domains, seniority_levels ["director", "vp"], current_functions ["Sales"], limit 30 — returns the buying committee across all of those accounts in one call for 9 credits.
- 4Run `find_work_email` on the named champion and economic buyer at each account, not the whole 30.
Outcome: A short account plan per qualified domain: the competitor in place, two named contacts with LinkedIn URLs, and a work email for each.
Champion tracking from your customer base
- 1Run `find_people` with past_company_domains set to your top customer domains, seniority_levels ["director", "vp"], recently_changed_jobs true, limit 25 — returns former users now in new seats for 7.5 credits.
- 2Run `enrich_person` on each LinkedIn URL to confirm the current job title and company, since the search snapshot can lag a very recent move.
- 3Run `enrich_organization` on the new company domains to check industry, size and funding, and drop accounts too small to buy.
- 4Run `find_work_email` then `verify_email` on the survivors to get a deliverable address at the new employer.
Outcome: A warm list of former users at new accounts, each row carrying the new company, current title, LinkedIn URL and a verified email for a referral-style opener.
Prompts
Paste any of these into Claude once the skill is installed and the SyncGTM MCP is connected.
Find VP Sales and Head of Sales at Privately Held companies in Software Development with 51-200 employees in the United States, and only people with a verified business email. Limit 25.Find c_suite and vp people in the Sales function in Germany who recently changed jobs, at companies with 201-500 employees. Limit 25, one page only, then tell me how many credits that cost.Find directors in the Sales function at stripe.com, shopify.com and figma.com using current_company_domains. Limit 20, and return name, title, company and LinkedIn URL in a table.Find people with Kubernetes or Terraform in their skills who studied at Stanford University and have at least 8 years of experience. Limit 15.Find people who used to work at hubspot.com and are now VPs at companies with 501-1000 employees in the United States. Limit 20, then find work emails for the 10 best fits and verify them.Best practices
Always pass at least one filter. A search with nothing to match on still bills 0.3 credits for every profile it returns, and the rows are noise.
Treat limit as the cost dial, not a page size, and cap the paging before you start. Three pages of 100 is 300 profiles and 90 credits, and cursor will keep going otherwise.
Use current_company_domains instead of current_company_names. Domains match exactly; company names are easy to get wrong and fail silently.
Use the exact enum strings. seniority_levels accepts c_suite, vp, director and senior — "C-level" and "Vice President" match nothing. industries needs the LinkedIn spellings from the reference list, so "Software Development" works and "SaaS" does not.
Pass the normalized headcount buckets: 1-10, 11-50, 51-200, 201-500, 501-1000, 1001-5000, 5001-10000, 10001+. The upstream Crustdata list shows "myself only" and "2-10" — those will not match the MCP.
Searching a single known company here is waste. find_people_within_company is a flat 1.5 credits and takes a domain.
Filters and accepted values
Filters only match when the value is spelled exactly as the source list has it. Point the agent at these lists when you build a query.
- Industries
Flat array of industry strings for the industries parameter, e.g. "Software Development", "Financial Services". Exact match only — "SaaS" and "Tech" are not in the list.
- Regions
Object keyed by region name to LinkedIn geo id, e.g. "Greater Minneapolis-St. Paul Area". Use these spellings for person_locations. Large file, around 4.5 MB.
- Job titles
Object keyed by title to id, e.g. "Co-Founder", "Senior Software Engineer". current_titles is fuzzy and any-of, but starting from a real title beats guessing.
- Skills
Flat array of skill strings for the skills parameter, e.g. ".NET Core", "Kubernetes". Any-of matching — one hit qualifies the profile.