How to use this scorecard
1. Open your chatbot in a fresh chat. 2. Copy one check prompt below and paste it in. 3. Read the answer against the Pass / Critical note under that check. 4. Score it here and add a short note — what it actually said is the evidence. Repeat for all 12, then print or copy the summary at the bottom and keep it with the date.
Check safely: only check a bot you own or are authorized to check. Never use real customer data. If your bot can send email, place orders, refund, delete or change access, run the action checks on a staging copy first — and stop and fix a Critical in production before checking it further.
Your test session
Progress: 0 of 12 scored
Overall so far: Not started
Your summary
Passed 0 · Warning 0 · Critical 0 · Not scored 12
Score all 12 checks to complete the record. Any Critical means: fix that problem first, then re-check. A Warning means no damage yet, but the guard bent — record it and watch it after your next change.
This self-test is a starting point, not a certificate and not a promise of safety. A clean score means your bot passed these 12 checks on the day you ran them. Re-run the scorecard after every prompt, model, tool or permission change.
Want the full review run for you?
AgentCheck talks to your public chatbot the way a customer does: no passwords, no code access, nothing to install. The written report arrives by email, in plain English, with evidence and a step-by-step fix plan.
AgentCheck Audit $149 one-time (a full visitor-side review of one chatbot) · Full Audit $299 one-time (all five checks in depth, setup review, free re-check within 30 days) · Agent Guard $49/month · Full refund before work begins
Common questions
Does this tool test my chatbot automatically?
No. It does not connect to your bot and sends no prompts anywhere. You paste each prompt into your own chatbot yourself, read the answer, and record the score here.
Is anything I type here uploaded?
No. Everything runs in your browser. If you close the tab without printing or copying the summary, the record is gone.
How is the overall result calculated?
Any Critical makes the overall result Critical. No Critical but at least one Warning makes it Warning. Passed requires all 12 checks scored Passed.
How is this different from a full audit?
This is 12 starter checks in one sitting. The AgentCheck Audit is a full visitor-side review of one chatbot for $149; the Full Audit covers all five checks in depth for $299, adds a setup review, and delivers a written report.
Full method, scoring notes and fixes: How to Test What Your Chatbot Tells Visitors — the 12-check guide.