Law prompt of the day

Deposition outline builder: topic modules, foundation questions, and impeachment setups from your anonymized record

You have a witness to depose and a hot-doc log or fact summary already built. Turn it into a working examination outline — topic modules, the exhibits to introduce, the admissions to lock in, and the impeachment setups — before you burn a weekend on it.

You are a litigation associate building an INTERNAL deposition outline for a licensed attorney's review — work product and a drafting aid, not legal advice, and not a substitute for the examining attorney's judgment. I will paste an ANONYMIZED case summary, the deponent's role, and excerpts or a hot-doc log (party and witness names replaced with role labels; Bates or exhibit numbers kept).

Produce:

A) EXAMINATION ROADMAP — the topic modules in the order you would take them (background and qualification first, then the substantive topics that build toward the admissions I need, sensitive topics last), each with a one-line objective tied to an element of my claims or defenses.

B) PER-MODULE QUESTION BLOCKS — for each topic: the foundation questions that establish personal knowledge and admissibility first, then substantive questions phrased to elicit a committed yes/no answer, then the exhibits to introduce (use ONLY the Bates/exhibit references I gave you) with the foundation each exhibit needs.

C) ADMISSIONS TO LOCK IN — the specific facts this witness can concede that advance an element or foreclose a defense, and the exact question that pins each one.

D) IMPEACHMENT SETUPS — where a prior statement I provided (deposition, declaration, document) conflicts with the witness's expected testimony: quote both sides from my inputs side by side and lay out the commit-then-confront sequence. Mark anything you inferred rather than read as "INFERENCE — confirm against the record."

E) ANTICIPATED PROBLEMS — likely objections (form, privilege, scope), evasive-answer patterns to expect, and the follow-ups that close each loop.

Inputs: [PASTE ANONYMIZED CASE SUMMARY + HOT-DOC LOG WITH BATES REFS] · [DEPONENT ROLE + RELATIONSHIP TO THE MATTER] · [MY CLAIMS/DEFENSES + THEIR ELEMENTS] · [THE ADMISSIONS I MOST NEED]

Rules: Anonymize before you paste — replace party, witness, and third-party names with role labels and keep privileged material, work-product strategy notes, and protected personal data (SSNs, medical details, account numbers) out of a consumer LLM entirely (ABA Formal Op. 512: informed consent is required before client confidences enter a self-learning tool). Do not invent testimony, prior statements, exhibit contents, deposition transcripts, or case law — build only on the excerpts and facts I provided, and where a question assumes a fact not in my inputs, flag it "RECORD CHECK: [what to confirm]" instead of asserting it. Verify every exhibit reference, every quoted prior statement, and every date against the actual record before the outline reaches a deposition, and confirm your jurisdiction's deposition rules and any protective-order limits yourself — the examining attorney owns the strategy, the objections, and the questions actually asked.

Why it works — A deposition outline fails two ways — it wanders instead of building toward the admissions that win the motion, and it lets the witness's prior inconsistent statements slide by unconfronted. Organizing by topic module with a foundation-first question order fixes the first; commit-then-confront impeachment blocks that quote both statements from the record fix the second. And because AI depositions get dangerous when the model invents testimony or an exhibit that was never produced, every fact chains to a Bates reference the attorney can pull and any assumed fact surfaces as a RECORD CHECK — never as a fabricated quote.