Email finder API for AI agents
Find a work email from a name and a company domain, verify it, and enrich people and companies, from Claude, ChatGPT or code. In two 150-lookup tests it found 86% and 94% of work emails at about 2.5 cents each, and a lookup that finds nothing costs nothing.
Lookups only. These tools find, verify and enrich contact data; they never send an email or a message, and they never touch an inbox of yours.
updated Nov 3, 2026
Copy the setup prompt.
Your agent does the rest.
Install superagnt for me. It's the superagnt plugin from the github.com/superagnt/plugins marketplace, plus the superagnt MCP server (on Claude it's a connector I add in my browser; elsewhere I sign in through my browser). The setup guide for each kind of agent is at https://www.superagnt.com/install» the lead tools are part of every server's base tools, so a fresh connection already has them.
How do you choose an email finder API?
Two numbers decide it: how many of your names come back with an email you can send to, and what each usable email costs once misses are counted. Then check what happens on a miss and whether you can test the API before paying. Ours, measured on 150 people, against a popular per-credit finder:
| superagnt | Snov.io | |
|---|---|---|
| Find rate on a stated sample | 129 of 150 from team pages (86%), 141 of 150 from people search (94%), Nov 3, 2026 | Not published |
| Cost per found email | 2.46 to 2.48 cents measured (2.45 to 4.97 cents a lookup) | About 3.9 cents a credit on Starter ($39 for 1,000 credits) |
| A lookup that finds nothing | Free | Free; an unverifiable "unknown" email still costs a credit |
| Verification status on each result | Yes: valid or accept_all (catch-all) | Yes: green, yellow or red |
| Budget cap per call | Yes, max_cost_cents | No |
| Test the API before paying | Yes: $3 of free credit, no card, API from the first call | Free Trial has no API access |
| Also in the same key | LinkedIn, X, Reddit, web, SEO data, a database and schedules | Email outreach tools |
checked Nov 3, 2026 · Snov.io pricing; What is a credit (Snov.io knowledge base, updated Oct 20, 2026)
What can your agent find and enrich?
A name plus a company domain (or company name) returns the work email with its verification status.
Check deliverability before you send, so a sequence doesn't burn a sending domain on bounces.
Job title, seniority, company, location, work email and profile links, from an email or a LinkedIn URL.
Firmographics from a domain: industry, headcount, location, founding year and description, plus funding, competitors and technographics.
Find people or companies by title, seniority, department, industry, headcount and location, 25 people a page.
The public addresses at a domain, each marked personal or role inbox, and bulk enrichment of up to 100 people or companies in one call.
data_agnt_contactsaction: find_email · verify_email · domain_emailsdata_agnt_peopleaction: search · enrich · bulk_enrichdata_agnt_people_find_mobilecalled by its own namedata_agnt_companiesaction: search · discover · enrich · bulk_enrich · intelligence
per-operation names12older servers list these
data_agnt_people_email_finderdata_agnt_people_email_verifierdata_agnt_people_enrichdata_agnt_people_find_mobiledata_agnt_people_searchdata_agnt_people_bulk_enrichdata_agnt_companies_discoverdata_agnt_companies_domain_emailsdata_agnt_companies_enrichdata_agnt_companies_searchdata_agnt_companies_bulk_enrichdata_agnt_companies_intelligenceWhat does a real call return?
A run we recorded on superagnt, with the result trimmed to fit. Nothing on it is invented: the date, the wall time and the cost are the run's own.
asked List the public email addresses at superagnt.com and tell me which are role inboxes.
- → call
data_agnt_contacts{ "action": "domain_emails", "domain": "superagnt.com", "limit": 5 }← resulttrimmeddomain: superagnt.com · disposable: false · webmail: false · accept_all: false | value | type | confidence | department | | security@superagnt.com | generic | 75 | support | | legal@superagnt.com | generic | 75 | legal | 2 emails
the same call over httpcurl · python
curl -X POST "https://api.superagnt.com/v1/data/agnt/companies/domain-emails" \
-H "Authorization: Bearer $SUPERAGNT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"domain":"superagnt.com","limit":5}'import os, requests
r = requests.post(
"https://api.superagnt.com/v1/data/agnt/companies/domain-emails",
json={"domain": "superagnt.com", "limit": 5},
headers={"Authorization": f"Bearer {os.environ['SUPERAGNT_API_KEY']}"},
)
print(r.json())» We ran this one on our own domain: the find test below used 150 real people, and its page shows only totals.
Which email finder API is the most accurate? Two tests, with their samples
Two tests on Nov 3, 2026 asked the finder for 150 work emails each, from first name, last name and company domain, with no budget cap. On names taken from companies' own team and about pages it found 129 (86%) at 2.48 cents per found email. On names drawn from the people search it found 141 (94%): 136 verified valid and 5 at catch-all domains, at 2.46 cents per found email. Misses cost nothing in either test.
| Where the 150 names came from | Found | Valid | Catch-all or unknown | Not found (free) | Cents per found email |
|---|---|---|---|---|---|
| Companies' own team and about pages | 129 | 123 | 6 | 21 | 2.48 |
| superagnt's people search | 141 | 136 | 5 | 9 | 2.46 |
- By sample of 50: on team-page names, US software staff 48, UK and EU small-business owners 36, e-commerce founders and staff 45; on people-search names, US B2B software staff 48, UK and EU small-business owners 46, US e-commerce marketers 47.
- The finder beats guessing: for the same 50 US team-page names, trying first.last@ and flast@ through the verifier found a valid address for 29 of them; the finder found 48.
- Its "valid" holds up: re-verified, 122 of the 123 team-page finds marked valid checked valid again.
- In the people-search test no result was a role inbox such as info@ or support@: the finder returns a person's own address.
- Median time per lookup: 6.6 and 7.2 seconds in the two tests.
sample · Team pages: 150 names from companies' own team and about pages, 16:55 to 17:27 UTC. People search: 150 people from superagnt's people search, 16:37 to 16:43 UTC, which marks whether an email is on file (it was for 144 of the 150), so that rate is what a search-then-find workflow gets. Both Nov 3, 2026
What does a lookup cost, and when is it free?
A find costs 2.45 cents when the first source has the email, and goes on to a second source only when the first misses, up to 4.97 cents. A lookup that finds nothing costs nothing, and so does a refused or failed call. Set max_cost_cents on a call to stop the waterfall at a price you choose: at the 2.45-cent floor, the same 30 people returned 28 emails against 29 uncapped, at 2.45 cents each.
| Call | What you send | Measured cost |
|---|---|---|
| find_email | Name plus domain or company | 2.45 cents a found email, up to 4.97 |
| verify_email | An email address | 0.2 cents |
| domain_emails | A domain | 2.52 cents |
| people enrich | An email or a LinkedIn URL | 1.98 cents, up to 6.95 |
| people search | Filters | 1.98 cents a page of 25 |
sample · Prices measured on the calls in this test, Nov 3, 2026
Can I find an email from a LinkedIn URL?
Yes, in one call: people enrich with the LinkedIn URL returned the work email for 10 of the 10 people we tried, at 1.98 cents each, cheaper than a name-and-domain find. The finder itself takes a name and a domain, so when all you hold is a profile URL, enrich first.
sample · 10 LinkedIn profile URLs from the same samples, Nov 3, 2026
Find, verify or enrich: which do you need?
Find turns a name and a company into an address. Verify checks an address you already hold before you send to it. Enrich turns an email or a LinkedIn URL into a person record: title, company, location and links.
- A catch-all (accept_all) domain's mail server accepts every address, so no check can confirm one person's mailbox. Treat those finds as unconfirmed: re-verified, 4 of 5 catch-all finds in the team-page test came back invalid. Verify before you send.
- Role inboxes (info@, support@) come back marked "generic" from domain_emails, so an agent can keep them out of a sequence.
- Send the domain when you have it. For the same 50 US people the finder found 48 by domain and 45 by company name alone, and 3 of those 45 landed at a different domain than the company's site.
What do people use it for?
Turn a name into a sendable contact
The agent takes a first name, last name and domain, finds the work email, checks its status, and only then writes the contact to your sequencing tool.
A budget-bounded list build
For each row the agent asks only for the fields it needs and sets max_cost_cents, so a 5,000-row build has a ceiling you chose rather than a bill you discover. Rows land in the workspace database.
Account research before a call
Given a domain the agent enriches the company, pulls funding, competitors and technographics, and adds what LinkedIn says about the team, in a one-page brief for the rep.
What does it cost?
Heavier calls cost more; each response reports its own cost.
About 1,500 calls at the lowest price.
Only a call that returns data is billed: a miss, a refusal or an upstream failure costs nothing. One credit balance covers this source and every other one, so there is no per-source subscription and no upstream account of yours to keep alive. After the trial, plans are Starter $49 a month ($20 of data credit included), Growth $149 a month ($90) and Scale $399 a month ($300). Requests per minute: Starter 60, Growth 300 and Scale 600.
What do people ask before they wire it up?
01Which email finder API is the most accurate?
Judge one by its find rate on a sample like yours and by how many found emails verify. Ours found 129 of 150 work emails (86%) on names from company team pages and 141 of 150 (94%) on names from a people search, in two tests on November 3, 2026, and 122 of 123 "valid" finds re-verified valid. Most finders publish no such number.
02What does an email finder API cost, and is there a free one?
Here a found email costs 2.45 to 4.97 cents, 2.46 cents on average in our test, and a miss costs nothing. There is no free plan, but the no-card trial gives $3 of credit, enough for about 120 finds, with the API on from the first call.
03Am I charged when no email is found?
No. A lookup that finds nothing costs nothing: the 9 misses in our 150-lookup test cost $0. Refused and failed calls are free too.
04Can I test before paying?
Yes. Sign up, verify your email and the trial gives $3 of data credit with no card. Your agent can call the finder in Claude, ChatGPT or Codex, or you can call it over HTTP, straight away.
05Can I find an email from a LinkedIn URL?
Yes. Call people enrich with the LinkedIn URL: in our test it returned the work email for 10 of 10 profiles at 1.98 cents each. The finder itself needs a name and a company domain.
06What is a catch-all domain and why does it matter?
A catch-all domain's mail server accepts mail for any address, so a verifier can't confirm that one person's mailbox exists. Results from such domains come back as accept_all instead of valid. Treat them as unconfirmed: in our test, 4 of 5 catch-all finds came back invalid when re-verified.