FlowRunner
PricingContact
Theme
Start Free

Kickbox

Email

Protect sender reputation before a campaign goes out. Agents verify email addresses in real time with Sendex quality scores, run batch verification over large lists, download results as CSV, catch typos with suggestions, and watch remaining credits.

5 actions API key available
Kickbox website ↗ Platform Documentation ↗ Capability data verified 2026-07-27
A campaign list of 40,000 addresses is staged for next week's send
Get Balance confirms the credit pool covers the batch
The marketing ops lead approves the credit spend for the full-list verification
Start Batch Verification submits the newline-delimited list and returns a job id
Get Batch Verification polls until the job reports complete
Download Batch Results pulls the per-address CSV with results, reasons, and flags
Deliverable addresses pass through; undeliverable ones are suppressed; risky and accept_all addresses queue for a human call before the send

What This Integration Enables

Sender reputation is an asset you spend every time you hit send, and an unverified list is how you spend it fastest. Mailbox providers grade the sender, not the campaign, so one careless blast taxes every send that follows it. Kickbox tells you, per address, whether the send will land: deliverable, undeliverable, risky, or unknown, with a Sendex quality score and flags for role, free, disposable, and accept_all addresses. The FlowRunner connector makes that check a standing part of every workflow that collects or mails an address. - Verify addresses in real time at signup, with typo suggestions surfaced while the person is still on the form - Clean large lists in bulk with asynchronous batch jobs, polled to completion and downloaded as CSV - Filter disposable, role, and accept_all addresses before they reach your marketing tools - Watch the credit balance so a big job never dies halfway The Sendex score is what separates this from a boolean bounce check. Deliverable is not a single grade: a deliverable address at a disposable domain, a role account like info@, and an engaged personal address are three different marketing decisions, and the flags let workflows treat them differently instead of mailing all three the same way. Verification itself is read-only. The two decisions around it, spending credits on a large batch and deciding what happens to the risky tier, are where the [human-in-the-loop](/concepts/human-in-the-loop) steps belong.

Without FlowRunner

Bounces discovered mid-campaign The list's decay shows up as a bounce spike, and the sender score pays for it
Typos become dead subscribers gmial.com signups enter the list broken and never receive anything, ever
List cleaning as an annual event Verification happens when deliverability is already damaged, which is the expensive time to learn

With FlowRunner

Every address verified at the door Signups get Verify Email in real time, with did_you_mean catching typos before they persist
Campaigns send to verified lists Batch verification runs before the send, and only deliverable addresses ride
Credits and risk managed deliberately Balance checks and spend approvals happen before jobs start, not after credits run out

Use Case Scenarios

The signup form that catches its own typos

A [Typeform](/integrations/typeform) submission arrives. The agent runs Verify Email before the address goes anywhere. Deliverable addresses sync onward. An undeliverable address with a did_you_mean suggestion triggers a correction prompt. Disposable addresses are declined for the gated content flow. The CRM only ever receives addresses that can actually be reached, and the correction prompt catches the typo while its owner is still there to fix it.

Pre-campaign list hygiene, end to end

Before the quarterly campaign, the agent pulls the list from [Google Sheets](/integrations/google-sheets), checks Get Balance, and posts the address count and credit cost for approval. On approval it runs Start Batch Verification, polls Get Batch Verification, and processes Download Batch Results: deliverable rows write back as the send list, undeliverable rows move to a suppression tab, and risky rows post to [Slack](/integrations/slack) with their Sendex scores for the marketing lead to keep or cut. The campaign in [Mailchimp](/integrations/mailchimp) sends to a list that was verified this week, not last year, and the verification CSV files away as the evidence of when the list was last cleaned and why each cut was made.

New customers scored before they enter the flows

When a new customer appears in [Shopify](/integrations/shopify), the agent verifies the address and routes on the result. Deliverable customers enter the post-purchase sequence immediately. Risky and unknown results hold in a review queue rather than dragging down the sender score, and accept_all corporate domains get flagged so the lifecycle team knows engagement data will be the only real signal there. Over time the review queue itself becomes data: a rising share of risky signups from one acquisition channel says something about the channel, and the weekly summary surfaces it.

Human-in-Loop Highlight

Start Batch Verification prices every address it checks against a prepaid credit pool, and Get Balance exists precisely because that pool runs dry at inconvenient times. A 40,000-address job is a real spend, and a retry loop that resubmits a list is the same spend twice. So the FlowRunner agent quotes before it commits: balance, address count, and net cost post to the marketing ops lead, and the job starts on approval. The second human moment comes after the results land. Undeliverable is not a judgment call, but risky and accept_all are: cutting them protects the bounce rate, keeping them protects the pipeline, and the right answer depends on the campaign. The agent sorts, scores, and stages; the person decides who gets cut from the list. Suppression is cheap to apply and expensive to un-apply once the send has gone out without them.

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

5 actions

Verification

4
  • Verify Email Verifies a single address in real time, returning deliverable, undeliverable, risky, or unknown with a reason code, a Sendex quality score from 0 to 1, flags for role, free, disposable, and accept_all, and a did_you_mean suggestion for likely typos. The at-the-door check for signups and single records.
  • Start Batch Verification Starts an asynchronous job verifying a newline-separated list of addresses, returning a batch job id. The bulk cleaner for large lists, and the credit spend this page's approval gate prices before it runs.
  • Get Batch Verification Retrieves a batch job's status by id, reporting progress while running and per-address results once finished. The polling step between submission and download.
  • Download Batch Results Downloads a completed batch job's full results as CSV, one row per address with result, reason, and quality flags. The artifact the routing logic and the audit trail both work from.

Account

1
  • Get Balance Returns the verification credits remaining on the account. The pre-flight check that keeps a large job from dying at 60 percent.

Frequently Asked Questions

What can FlowRunner do with Kickbox?

FlowRunner agents can run Verify Email, Start Batch Verification, and Get Batch Verification in Kickbox, plus 2 more actions.

Does connecting Kickbox to FlowRunner require OAuth?

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

Can Kickbox trigger a FlowRunner workflow automatically?

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

Start building with Kickbox

$100 in credits. No card required. Connect in minutes.