Pending isn't paid. Free 2-minute check
Find the money-answer mistakes your AI agent may be making.
Five money questions your customers ask. Pick the reply closest to what your agent would say, or tap Not sure. You'll see the pattern behind each mistake and what a safer answer does.
Each pattern here showed up as a Critical failure when I tested an AI agent for AcmeSend, a made-up transfer company. This is a quick pattern check, not a live test of your agent.
Your results
These are your picks, not a test of your agent.
Free
Get the five corrective examples
Paste-ready wording for each pattern. Enter your work email and they unlock here.
Written for a typical status model. Use your own status names and policies when you paste them in.
My estimate: writing and grading 50 scenarios like these takes your team about 17 hours, at about 20 minutes each. With the Score, I do that work for you.
Where these come from
In my AcmeSend demo, a real AI agent failed 8 of 25 money questions at Critical. After a one-page fix, none of those 8 did. Read the AcmeSend report
Who's reading your answers
Allana Jackson. I write the instructions and tests AI support agents run on. Before that: Google's Search Ads 360 billing and payments help center, plus knowledge systems for DoorDash and Domino's.