This page lists three turns, the messages you send to the AI one after another in the same chat, and then a check of the answer. Keep it open next to your chat. Each turn shows the next thing to paste, with the values you entered already filled in. Nothing is sent from this page, and your conversation stays in the AI tool you are using.
State it as the question the court has to answer, not as a topic.
The three to six facts that should drive the analogy. Note which are in the record and which are contested.
The controlling court, and whose substantive law applies if it differs.
Where this gets decided. The standard changes which cases actually help.
Cases you are relying on and, just as usefully, the ones that hurt you.
Watch for these throughout
Models fabricate case names, reporter cites, and holdings that read perfectly. Pull and Shepardize every citation before it touches a filing. Courts have sanctioned lawyers for exactly this.
A case can be good law and still cited for the wrong proposition. Confirm the pin cite says what the parenthetical claims.
Subsequent history moves. A case that was strong last year may have been limited or abrogated since, and the model cannot see that.
Unpublished and out-of-jurisdiction decisions carry citation restrictions in many courts. Check the local rule before relying on one.
Analogies built on contested facts collapse at summary judgment. Confirm each driving fact is actually in the record.