Free scorecard · 10 questions

Does your AI learn? A 10-question learning-loop scorecard.

In a few minutes, see whether your AI keeps any memory between sessions, who can change it, and whether a lesson can become policy without a person.

An operator reviews a checklist of memory controls beside an AI assistant.

A lesson is not policy until a person approves it.

Built for B2B operators$5M to $100M revenueHuman review stays in the loop

A starting point, not a certification.

Choose the answer closest to how your system works today. The score is based on your answers, not an independent audit. Submit your details to see the two or three priorities to review with a person.

R1 to R4 governs actions. This scorecard checks the memory, ownership and promotion rules behind them. It complements the existing diagnostic and does not replace it.

A useful baseline

Answer for the system you have today.

Choose the closest answer. Your score is a self-assessment, not an independent audit or certification. Submit your details to see the result returned by the server.

10 questions Choose one answer for each.
01After a session ends, does any lesson persist outside the chat?
02Can you list every store that holds prompts, memory, or retrieved documents?
03How often does each of those stores update?
04Who may change a prompt, a memory record, or a retrieval source?
05Is that change reviewed before it affects production?
06Do you version memory separately from the model?
07For a past AI decision, can you show which version of the model and the memory produced it?
08What would trigger a re-validation of a workflow after it launched?
09If a bad lesson got into long-term memory, how would you roll it back?
10Is anyone named as the owner of what the agents are allowed to remember?

Where should we save your result?

Your details and any answers are sent with your request and are not saved in this browser. Do not include passwords, access keys or confidential data.

Keep the useful context. Keep the approval rule.

Start with what your system does today.

Map what changes, who owns it, and what needs a human decision before the next version is in use.