Mailcheck

Mailcheck verifies email addresses and cleans lists. Agents check a single address for validity, a disposable flag, catch-all and mailbox existence, and a trust rate before sending to it or storing it.

Verified 1 action API key available

Mailcheck website Platform Documentation Capability data verified 2026-08-19

A demo request or trial signup posts an address to the agent

Check Email returns the trust rate, the delivery signals, and the person signals

Agent separates what the check proved from what the domain made unprovable

Agent flags catch all domains, where a passing SMTP answer proves nothing either way

Qualified records are written to the CRM with the signals attached to the record

The owning rep is alerted when a signup carries a strong real-person signal

Sales ops decides the disposition of low scores that sit behind a catch all domain

What This Integration Enables

Most email verifiers answer one question: will a message sent here be accepted. Mailcheck answers a second one alongside it, and that is the reason to pick it. Its data comes from public sources, search, open databases, and social networks, with manual moderation layered on top of the automated checks, which is the vendor's own stated difference from a purely SMTP-based verifier. So a result carries a trust rate from 0 to 100 and the usual MX and SMTP evidence, and it also carries whether a Microsoft account is attached and whether a GitHub account is. The second kind of signal says nothing about deliverability. It says something about whether there is a person on the other end.

That matters most at the front of a pipeline. A free-trial signup from an address with an attached developer account and a clean trust rate is a different object from a syntactically valid address nobody has ever used anywhere. FlowRunner agents run Check Email at the moment a record is created, write both kinds of signal onto the record, and route accordingly. The connector adds positively named flags alongside the vendor's own, because three of Mailcheck's fields are phrased as negations and reading one backwards keeps exactly the addresses you meant to drop. Where the result is genuinely ambiguous, a human-in-the-loop step routes the case to sales ops instead of letting a threshold decide it.

Without FlowRunner

One number decides everything A trust score below the line rejects the record, whatever produced the number

Real people and real mailboxes get conflated An address that accepts mail says nothing about whether a person is behind it

Enrichment happens in a second tool Deliverability comes from one vendor and identity signals from another, joined by hand later

With FlowRunner

The score arrives with its reasons The individual checks travel with the verdict, so a decision can be made on the cause

Person signals sit on the record Attached accounts and the domain's own nature land in the CRM alongside the address

Ambiguous cases go to a person Records the check could not resolve are routed for judgment rather than silently dropped

Use Case Scenarios

Trial signups sorted by whether a person is behind them

A trial signup arrives. The agent calls Check Email and reads two things at once. The delivery evidence tells it whether a confirmation mail will land. The attached-account signals tell it whether this address has a history in the world. A signup with a clean trust rate and an attached developer account is written to HubSpot and posted to the owning rep in Slack the same minute. One with a passing syntax check, no MX answer, and no attached accounts anywhere goes to a review queue rather than to a rep. The rep's day fills with the first kind.
Cleaning a list before it becomes a sequence

Before a prospecting list is loaded into a sequencer, the agent runs Check Email across it and sorts on the disposable and catch-all flags rather than on the score alone. Confirmed disposable addresses are removed. Addresses whose mailbox answered over SMTP and whose domain is not a catch all are cleared for sending. Everything else is held. The cleared set is written to Pipedrive with the trust rate stored on the record, so a rep looking at a stalled sequence later can see whether the address was ever provably real.
Enrichment that stays in its lane

A form submission from Typeform reaches the agent. Check Email returns a GitHub username attached to the address. The agent writes it to the lead record as a research note and does nothing else with it, because it is enrichment rather than validation and proves nothing about whether the mailbox accepts mail. The routing decision still comes from the delivery signals. What the enrichment buys is context for the rep who opens the record, not a shortcut past the check.

Human-in-Loop Highlight

The trap in this connector is not the score. It is the catch-all domain. Check Email reports whether the domain accepts everything, and when it does, an SMTP answer proves nothing about the specific mailbox, so the trust rate falls without the address being wrong. Catch-all configurations are common at exactly the kind of large employer whose people you most want to reach. A mechanical cutoff applied across the whole result set will therefore reject your enterprise buyers and keep your consumer ones, which is precisely backwards.

So the qualification agent splits the below-threshold cohort before it does anything with it. It posts: "Batch of 3,400 checked. 210 fall under the trust threshold. 148 of those sit on catch all domains, 61 of them at accounts already in the CRM as open opportunities. The remaining 62 failed on syntax or a missing MX record. Drop the 62 and route the 148 for manual review, or apply the threshold to all 210?" Sales ops answers. The agent runs the checks; a person decides what a domain's own configuration is allowed to cost you.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via preferred channel

Human decides

Agent resumes with decision

Agent Capabilities

1 actions

Verification

Check Email Runs the single address check and returns the whole record: a trust rate from 0 to 100, whether MX records are present, whether the mailbox answered over SMTP, whether the domain is a catch all, whether the address is disposable, and whether Microsoft or GitHub accounts are attached to it. The connector leaves the vendor's fields untouched and adds positively named equivalents for the three that are phrased as negations, so a flow branches on a flag that reads the way it behaves. Used at signup, before a sequence, and in any hygiene pass where the question is whether a real person is behind the address rather than only whether the mailbox exists.

Frequently Asked Questions

What can FlowRunner do with Mailcheck?

FlowRunner agents can run Check Email in Mailcheck.

Does connecting Mailcheck to FlowRunner require OAuth?

No. Mailcheck connects to FlowRunner with an API key, no OAuth flow required.

Can Mailcheck trigger a FlowRunner workflow automatically?

Mailcheck doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with Mailcheck

Free plan, no card required. Connect in minutes.

Start Building Free Talk to us