Prompt Library

ChatGPT Prompts for Paralegals (Research, Drafting, Discovery)

16 copy-paste prompts

20 copy-paste ChatGPT prompts for paralegals: legal research support, document drafting, discovery organization, court filing prep, and the litigation support work that drives matter outcomes.

In short: This page contains 16 copy-paste ready prompts, organized into 4 categories with a description and pro tip for each. The first 5 prompts are free instantly, no signup needed. Hand-curated and tested by the AI Academy team.

Louis Corneloup
By Louis Corneloup · Founder, Techpresso
Last updated ·Hand-curated & tested by the AI Academy team

Prompts get you started. Tutorials level you up.

A growing library of 300+ hands-on AI tutorials. New tutorials added every week.

Start 7-Day Free Trial

Document Drafting

4 prompts

Pleading Draft from Outline

5/16

✨ What it does

ChatGPT drafts a pleading from [Paste matter facts + cause of action] in [jurisdiction] format, with caption, jurisdictional allegations, numbered facts, counts, prayer, and signature block. Mark the boilerplate, then have the attorney review and sign before filing.

[Paste matter facts + cause of action]. Draft pleading per [jurisdiction] format requirements: caption, jurisdictional allegations, factual allegations numbered, claims for relief by count, prayer for relief, signature block. Boilerplate sections noted. Attorney reviews + signs.

💡

Pro tip: Pleading drafts = paralegal supports + attorney finalizes. Local rules + format vary by jurisdiction; always check court-specific requirements before final.

Discovery Request Drafting

6/16

✨ What it does

ChatGPT drafts [interrogatories / RFPs / RFAs] for [matter type], numbered and scoped to the issues, with definitions and a signature line instead of recycled boilerplate. Tailor any request that still reads generic, then send only the set the attorney approves.

Draft [interrogatories / RFPs / RFAs] for [matter type]. Output: numbered requests, scoped to matter issues, specific not generic, appropriate to civil rules, definitions section, signature line. Avoid boilerplate from prior matters; tailored requests get better answers.

💡

Pro tip: Boilerplate discovery = boilerplate objections. Tailored requests = harder to dodge. AI helps tailor; paralegal applies matter-specific knowledge; attorney signs off.

Discovery Response Organization

7/16

✨ What it does

ChatGPT organizes [Paste discovery requests] into a response plan: what you have, what to gather, privilege concerns, objections, and a proposed answer per request. Track it for the production index, then let the attorney finalize objections and wording.

[Paste discovery requests]. Organize for response: per request, what we have, what we need to gather, privilege concerns, objections to assert, proposed response. Track for production index. Attorney finalizes objections + responses.

💡

Pro tip: Discovery response organization saves attorney time. Paralegal preps + flags issues; attorney makes objection calls + signs verifications. Good prep = clean responses + reduced sanctions risk.

Deposition Outline

8/16

✨ What it does

ChatGPT builds a deposition outline for [witness in matter] with opening admonitions, background, foundation, key documents, probe areas, anticipated objections, and follow-ups. Hand it to the attorney to customize for the live exam; do not run the deposition from the raw outline.

Deposition outline for [witness in matter]. Output: opening (admonitions), background questions, foundation for substantive areas, key documents to use, areas to probe, anticipated objections, follow-up questions per topic. Attorney customizes for live deposition.

💡

Pro tip: Deposition outlines guide; live deposition adapts. Paralegal provides structured outline + key documents organized; attorney conducts. Good outline = attorney focused on testimony, not paper-shuffling.

Discovery + E-Discovery

4 prompts

Document Production Index

9/16

✨ What it does

ChatGPT builds a production index for [matter] with Bates numbers, descriptions, source or custodian, date, privilege flags, and redactions in a review-platform-friendly format. Keep the index current as documents go out so the production has a trail.

Build production index for [matter]. Output: numbered Bates-stamped documents, description per document, source/custodian, date, privilege flag, redactions noted. Format compatible with attorney review platform. Index = production traceability.

💡

Pro tip: Production index quality = matter defensibility. Sloppy index = sanctions risk + wasted hours later. Bates-stamped + custodian-tracked = professional standard.

Privilege Log

10/16

✨ What it does

ChatGPT drafts privilege-log entries for [withheld documents] with Bates range, date, author, recipients and affiliations, privilege claimed, basis, and a brief description that does not reveal privileged content. Review each line for court-defensible wording before you produce the log.

Privilege log entries for [withheld documents]. Per document: Bates range, date, author, recipients (with affiliations), privilege claimed, basis, brief description (without revealing privileged content). Court-defensible if challenged.

💡

Pro tip: Privilege logs scrutinized by opposing counsel + courts. Vague "attorney work product" without basis = waiver risk. Specific basis + careful description = defensible. AI helps draft; attorney reviews.

E-Discovery Search Term Negotiation

11/16

✨ What it does

ChatGPT proposes e-discovery search terms for [issue], including custodians, date range, file types, a hit-count estimate, and a refinement plan if the set is too broad or narrow. Negotiate the list with opposing counsel, then get it into the order if the court requires one.

E-discovery search terms for [issue]. Output: targeted search terms, custodians, date range, file types, expected hit count estimate, refinement strategy if too broad/narrow. Negotiate with opposing counsel + court order if needed. Proportional discovery.

💡

Pro tip: Search term negotiation = critical e-discovery moment. Too narrow = miss relevant docs (sanction risk). Too broad = expensive review. Balanced terms = proportional + defensible.

Document Review Coding

12/16

✨ What it does

ChatGPT builds a document-review coding scheme for [matter] with relevance categories, privilege tags, issue codes, hot-doc flags, confidentiality designations, and a QC layer. Brief reviewers on the scheme, then use the QC pass so coding stays consistent.

Document review coding scheme for [matter]. Output: relevance categories, privilege tags, key issues coding, hot doc flagging, confidentiality designations, QC layer. Consistent coding across reviewers = defensible production.

💡

Pro tip: Coding scheme set up wrong = re-review cost + production errors. Set up right = scalable + defensible review. Attorney sets standards; paralegal often manages execution.

Like these prompts? There are full tutorials behind them.

Learn the workflows, not just the prompts. 300+ easy-to-follow tutorials inside AI Academy — and growing every week.

Try AI Academy Free

Court + Practice Operations

4 prompts

Court Filing Cover Letter

13/16

✨ What it does

ChatGPT writes a court-filing cover letter for [document being filed + court] to the clerk, referencing the attachment, conformed-copy request, fee, return envelope, and a contact for questions. Print it with the packet, then confirm the fee and copies before you walk it in.

Cover letter for court filing: [document being filed + court]. Output: professional letter to clerk, document attached referenced, conformed copy request, filing fee enclosed, return envelope, contact for questions. Court clerk efficiency = your matter handled smoothly.

💡

Pro tip: Court clerks process hundreds of filings; clear cover letters = your filing accepted on first attempt. Sloppy filing = rejected + delayed + missing deadlines.

Pretrial Order Outline

14/16

✨ What it does

ChatGPT outlines a pretrial order in [jurisdiction] format with stipulations, contested facts, witness and exhibit lists with objections, pending motions in limine, and jury questions if they apply. You compile the lists; the attorney finalizes the order.

Pretrial order outline per [jurisdiction] format. Output: stipulations, contested facts, witness lists, exhibit lists with objections, motions in limine pending, jury questions if applicable. Attorney finalizes; paralegal compiles.

💡

Pro tip: Pretrial order = roadmap for trial. Paralegal compilation accuracy = trial efficiency. Errors in exhibit list or witness list = wasted argument time. Detail-oriented work pays.

Trial Notebook Index

15/16

✨ What it does

ChatGPT builds a trial-notebook index for [matter] with tabs for opening, each witness (direct, cross, exhibits), motions, jury instructions, and closing, plus exhibit-to-witness cross-references. Print the tabs, then keep the notebook current so counsel can work from it in court.

Trial notebook index for [matter]. Output: tabs for opening, witnesses (one per witness with subtabs for direct, cross, exhibits), motions, jury instructions, closing. Cross-reference exhibits to witnesses. Attorney works from notebook in court.

💡

Pro tip: Trial notebook quality = attorney effectiveness in court. Paralegal craftsmanship here = invisible but critical. Top paralegals known for trial notebook excellence.

Client Update Letter

16/16

✨ What it does

ChatGPT drafts a client update on matter status that is professional but readable, covering current status, recent developments, next steps, what you need from them, and billing context if it matters. Leave the attorney sign-off line, then send only after they sign.

Client update letter on matter status. Output: professional but accessible, current status, recent developments, next steps, what we need from them, billing context if relevant, attorney sign-off line. Paralegals draft; attorneys sign.

💡

Pro tip: Client communication = retention. Paralegal-drafted letters quality = attorney relief + attorney sign-off speed. Attorney drafting from scratch = expensive use of time.

Free tool

Prompt Optimizer

Turn a rough idea into a structured, professional AI prompt.

Try it free →

Frequently Asked Questions

For brainstorming + first-pass orientation: yes. For authoritative legal research: never alone. AI hallucinates citations. Verify against Westlaw / Lexis / official sources. Attorney signs off; paralegal supports.
Standard ChatGPT is NOT confidential. Don't paste client info, case strategy, work product. Attorney-client privilege at risk if pasted. ChatGPT Enterprise has zero retention; even then, firm policy may dictate. De-identify before paste.
Document review automating; basic doc drafting automating; legal research automating. Paralegals shift to: project management, complex tasks, attorney support requiring judgment, client communication. Paralegals using AI = more leveraged.
Local rules + format requirements vary widely; AI doesn't know court-specific rules reliably. Use AI for orientation + drafts; verify against local rules + court site before filing. Wrong format = rejected filing = missed deadline.
Document drafting first pass, research synthesis, discovery organization, deposition outlines, client letter drafting, trial prep checklists. Productivity multiplier across whole role; replaces no single function.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

7-day free trial. Cancel anytime.