Find and Verify Work Emails from One Agent Prompt

The economics

The old way vs. the treg.to way

	The old way	With treg
What you pay for	Apollo $59/seat + Hunter $34/mo + Lusha $49/mo, and a credit pool that expires monthly	One prepaid balance. A found email is $0.0245, and a miss costs nothing
Keys	One account, one login and one API key per provider, spread across machines and `.env` files	One treg token. Every tool in the catalog answers to it
Picking a provider	You use whichever tool the team bought last year	`catalog get` lists all eleven company-search providers with price, measured success rate and median speed
Commitment	Seat minimums and annual contracts before you know whether the data covers your market	No subscription. Test the coverage for cents, then decide
The workflow	Export from the prospecting tool, paste into the enrichment tool, upload to the verifier, reconcile three CSVs	One agent run: find, enrich, verify, deduplicate — in one pass

Instead of$142/mo Apollo + Hunter + Lusha, at list

You paid$0.0245 what our run actually cost

Try it

A real workflow

Copy this into Claude Code, Cursor, Codex or opencode

1Set treg up in your agent first time only

Paste this into the same agent. It installs the CLI, signs you in and registers the tools. One line, once — already set up? Skip to step 2.

setup

set up treg — https://treg.to/llms.txt

2Run the workflow

the prompt

Find 50 recently funded AI companies with 20–200 employees. Identify their
head of growth or VP of marketing and verify their work email.

What happens when you run it

The agent finds the companies. catalog_search "search companies by size and funding" returns eleven providers for that one job — and this is the sharpest price spread in the whole catalog:

Provider	Cost per company	Success rate	Median
Akta	free	100% (248 calls)	3.2 s
Hunter Discover	free	100% (24 calls)	1.7 s
The Companies API	$0.0019	100% (339 calls)	0.3 s
Lusha	$0.004992 per 25 results	not yet measured	—
LeadMagic	$0.025	70% (10 calls)	3.2 s
Apollo	$0.026 per page	not yet measured	—
Diffbot	$0.0299	100% (5 calls)	0.7 s
PDL	$0.38	not yet measured	—
Crunchbase	own key only	—	—

Same job. A 200× spread, and the cheapest option is free. Coverage differs, which is exactly why the price is not the only column — but nobody should pay $0.38 a record without knowing $0.0019 was on the table.

It finds the people. Two routes, and the cheaper one is usually the right one. If you want everyone at a company, hunter.companies.emails returns the known addresses at a domain in a single call — 100% success across 597 calls, 0.4 s median, the fastest endpoint in this whole workflow. If you want one named person, hunter.people.email.find costs $0.0245 and only charges when it finds someone (100% across 582 calls, 1.4 s). Enrichment fills thin profiles: Hunter $0.0245 (356 calls, 0.3 s), Apollo $0.026 (263 calls, 0.3 s), LeadMagic $0.025, Coresignal $0.392.

It verifies before you send — and this is the step that matters most. Verification is the single most-called enrichment endpoint on treg.to: 963 calls at 100% success. It runs at $0.00625 with LeadMagic or $0.01225 with Hunter. Finding an address is cheap; sending to a dead one costs you a domain reputation, which is why the teams already doing this at volume verify more than they search.

What comes back

what comes back

50 companies · 20–200 employees · funded in the window · pulled [date]

COMPANY          SIZE   LAST ROUND      CONTACT              TITLE              EMAIL          STATUS
<company>        84     <round, date>   <name>               VP Marketing       <email>        verified
<company>        142    <round, date>   <name>               Head of Growth     <email>        verified
<company>        37     <round, date>   <name>               Head of Growth     —              no match
...

SUMMARY   50 companies · 47 contacts identified · 41 emails verified · 6 no match
COST      $[from your run]   (misses were not charged)

Structure is illustrative. Values come from the providers, relayed unchanged.

Evidence

Proof — from one real run

Run on treg.to, 17 Aug 2026. Every figure below is from the Activity log of that run.

Field	Value
Providers considered	11 for company search, 3 for finding a work email
Providers selected	`hunter.x.discover-companies` (free) · `hunter.people.email.find`
Why	Discovery is free on Hunter and takes a plain-English brief. For the email, Hunter's finder carries one of the longest records in the catalog — 582 calls, 100% success, 1.4 s median — and it only charges when it finds one
Total cost of the run	$0.0245 — company discovery was free, one email was found and charged
Subscription cost avoided	$142/mo at list — Apollo $59/seat + Hunter $34 + Lusha $49
Time to completion	Under 5 seconds across both calls
Data freshness	The address came back verified `valid` on 2026-08-17, sourced from a page last seen 2026-07-30
Companies returned	9, all matching "AI infrastructure, recently funded, 20–200 employees"
Emails found and verified	1 of 1 attempted, confidence score 80

What the discovery call returned, for $0.00: nine companies with their domains and a count of known addresses at each — Daloopa (223), ZincFive (64), Ethernovia (51), RunPod (33), AttoTude (19), Netris (17), Bobyard (16), Normal Computing (9), Arycs Technologies (4). Hunter resolved the plain-English brief into explicit filters and returned them: headcount bands 20-50 and 51-200, funding series pre-seed through series C+.

One call in this run failed and cost nothing. A company-search request was rejected for sending a boolean as a string — HTTP 400, $0.00 charged. Your parameter mistakes are free; only answers are billed.

The honest read of this run: one verified email is a demonstration, not a benchmark. Hunter's 100% over 582 calls is the measured success rate of the endpoint, not a promise that it finds every person you ask for. The number that matters for your list is the hit rate on your market, which is what the free tier is for — and misses cost nothing either way.

Outcomes

Three things you can do the day you sign up

Build a list for a campaign you have not committed to yet.

Test whether your ICP even exists at the size you assumed. Company search starts free, so you can check the market before anyone signs a contract for the data.

Enrich the leads you already have, without a new seat.

Point the agent at your existing CSV of names and domains. It fills in titles, company size and verified emails, and marks the rows it could not resolve rather than guessing.

Stop paying for bounces.

Verify every address before it enters the sequence. At $0.00625 a check, verifying a 2,000-row list costs about $12.50 — and Hunter's finder does not charge you for the ones it cannot find.

Questions

Before you sign up

Why not just call the providers directly?

Because this workflow is not one provider. Finding companies, identifying people and verifying email are three different products, and the vendor that is best at one is rarely best at the others — the table above is 200× wide for a single job. Holding accounts with all of them to find out is the expensive way to learn it. If you already pay for one, connect it and those calls route through your key, unmetered. treg.to is closer to OpenRouter for agent tools than to a data vendor: one base URL, one token, many providers behind it.

How are credentials handled?

The credential is injected on the server. Your agent makes the real upstream request through treg.to's /call/ endpoint; treg.to adds the key and relays the provider's answer back verbatim. No provider key is ever written to your machine, your repo or your agent's context. Every call is recorded and attributed to the token that made it.

Can I choose a specific provider?

Yes. Every endpoint has an id, and calling it by id calls that provider. If you want the choice made once for the whole team, treg org pin <capability> --provider <provider> refuses calls to any other provider of that capability. (Often the right call here. Coverage varies by region and company size far more than it varies by price, so once you find the provider that covers your market, pin it.)

Can I use my existing provider key?

Yes, and it takes precedence. Register the key once and every call to that provider routes through it — those calls are never metered against your treg.to balance. Your key always wins over treg.to's. (If your team already pays for Apollo or Hunter, this is the first thing to do. Those calls stop costing you anything on treg.to.)

What happens if a provider fails?

treg.to does not silently reroute you. That is deliberate: only you know which inputs you actually hold, so treg.to relays your request rather than rewriting it. What your agent gets instead is the information to recover — it already knows the alternatives and their parameters, so on a 429, a 5xx or a timeout it can try the next provider and tell you which one it switched to. Failed calls are not billed. If a call succeeded upstream but the answer was lost coming back, an Idempotency-Key returns the stored result without paying twice.

How much does a call cost?

A found work email is $0.0245 with Hunter and a miss is free. Verification is $0.00625. Company search ranges from free to $0.38 a record depending on provider. The exact price is shown before the call, and treg.to adds no markup to what the provider charges. New teams start with $1.00 of free credit — enough for roughly 40 verified email lookups before you spend anything.

Which agents does it work with?

The installer registers treg.to's MCP server into Claude Code, Cursor and opencode automatically. Any MCP client that supports the authorization spec can connect with OAuth. Anything that can run a shell command — Codex included — can use the treg CLI or plain HTTP.

Find and Verify Decision-Makers from One Agent Prompt