
Kickbox
Every time a visitor shares their email, your AI agent verifies it through Kickbox in real time. Disposable domains get flagged, typos get caught, and your list stays clean. No more bounced campaigns or wasted send credits because an invalid address slipped through your forms.




Your AI agent validates email addresses the instant a customer types them, using Kickbox's verification engine to protect your sender reputation before bad data enters your system.
Kickbox
Real scenarios where AI-powered email verification happens during the conversation, not after, ensuring every address captured is worth a follow-up.
A visitor fills out your chat lead form with a random string at a fake domain. Your AI Agent sends the email to Kickbox's verification API, gets back an 'undeliverable' result, and immediately asks for a valid address. The visitor corrects it to their real email. Your sales team only sees verified leads in the CRM, and your outbound campaigns never bounce.
A new customer types 'jane@yaho.com' during account registration through your AI chatbot. Kickbox returns a did_you_mean suggestion of 'yahoo.com.' The agent responds: 'Looks like there might be a typo. Did you mean jane@yahoo.com?' One click and the customer proceeds with the right address. No lost activation emails, no support tickets about missing confirmations.
Your marketing team runs a webinar and uses your AI agent to handle registrations. For every email submitted, the agent checks Kickbox's Sendex quality score. Addresses scoring below 0.5 get flagged as low quality, and the visitor is asked to provide a work email instead. Your webinar list fills with high-quality contacts, and post-event nurture campaigns see higher open rates.

Kickbox
FAQs
When a visitor provides their email, the agent sends it to Kickbox's Single Verification API in real time. Kickbox checks SMTP deliverability, domain validity, and mailbox existence, then returns a result (deliverable, undeliverable, risky, or unknown) along with a quality score and flags. The agent uses this response to either accept the email or ask for a correction.
Sendex is Kickbox's proprietary quality score ranging from 0 (lowest quality) to 1 (highest quality). It evaluates factors beyond simple deliverability, such as engagement history patterns. You can configure the agent to accept only emails above a minimum Sendex threshold, like 0.6, ensuring your list quality stays high.
Yes. Kickbox returns a did_you_mean field when it detects a likely typo in the email domain. For example, if someone types 'user@gmial.com,' the API suggests 'gmail.com.' The agent presents this suggestion to the visitor conversationally and lets them confirm or re-enter their address.
Tars passes the email to Kickbox for verification during the conversation and uses the result to proceed. Email addresses entered during conversations are handled according to your Tars data retention settings. Kickbox's own retention policies apply on their end, and they do not share verification data with third parties.
Kickbox's single verification typically responds within 1 to 3 seconds. The SMTP verification timeout defaults to 6,000 milliseconds but can be configured up to 20,000 milliseconds via the timeout parameter. In practice, customers experience a brief pause that feels natural within a conversation flow.
Yes. Kickbox's response includes a 'role' boolean flag that identifies addresses like info@, support@, and admin@. The agent can be configured to accept role-based emails with a note, reject them entirely, or ask for a personal address depending on your lead qualification rules.
An 'unknown' result means Kickbox could not definitively verify the address, often due to a catch-all mail server or temporary SMTP issues. You can configure the agent to accept unknowns with a lower confidence flag, retry verification, or ask the visitor to confirm their address is correct.
Backend validation happens after form submission and often shows a generic error page. With Tars, verification happens inside the conversation. The agent explains the issue in plain language, suggests corrections, and collects a valid address without the visitor ever leaving the chat. It is a smoother experience with higher completion rates.
Don't limit your AI Agent to basic conversations. Watch how to configure and add powerful tools making your agent smarter and more functional.

Privacy & Security
At Tars, we take privacy and security very seriously. We are compliant with GDPR, ISO, SOC 2, and HIPAA.