What ReplyAuditor is

ReplyAuditor is a quality gate for AI-drafted customer support replies. You paste a reply, optionally add brand or policy rules, pick a channel, and get a quality report that checks hallucination, PII leaks, brand tone, and policy compliance — ending in a ship/hold verdict. It is built for support team leads deploying AI to draft responses.

What it checks

  • Hallucination — flags unverifiable facts: order numbers, tracking IDs, refund amounts, delivery dates.
  • PII leaks — flags credentials, national IDs, card numbers, and other personal data in the draft.
  • Brand / tone — checks the reply against the rules you supply.
  • Policy compliance — checks the reply against your support policies.

What the ship/hold verdict means

The verdict compresses the checks into one signal. Hold means at least one check failed and a human should look before sending. Ship means it passed the gate's checks — not that it is guaranteed flawless. The verdict is a prompt for human judgment, not a substitute for it.

Does it guarantee compliance?

No. ReplyAuditor flags potential issues and returns a verdict; it does not guarantee a reply is accurate, compliant, or risk-free, and it provides no legal sign-off. The deploying team remains responsible for what goes out. Treat it as a seatbelt, not an autopilot.

How it is priced

  • Free — a limited number of replies per month.
  • Pro — a monthly price with unlimited audits and tone rules.

The checks are the same across tiers; the difference is volume and the team QA dashboard.

Where to go next

For a first walkthrough, see the get-started guide. For the failure modes it catches, see common mistakes when shipping AI replies.

References

  • GDPR — personal data in customer communications: https://gdpr.eu/