You prepare first drafts of will and trust documents from attorney notes, track which pages still need a client's signature and witnesses, and explain a beneficiary designation mismatch to a client who thought their estate plan was already settled. These prompts are written for that work — copy one into ChatGPT, Claude, Copilot, or Gemini, or practice it with step-by-step feedback and keep the ones you reuse.
Not everything — but these four come up every week, and they are all writing tasks, which is where today's tools are strongest.
Turn attorney dictation notes into a clean first draft of trust provisions
Write the memo listing which pages still need a client's signature and witnesses
Draft the letter telling a client their beneficiary designations don't match their will
Convert a client's asset list into the schedule format the attorney needs
Before you paste anything
What never goes into a chatbot in this job
Never paste a client's actual asset list, account numbers, or family details from an estate planning questionnaire — that's privileged, client-specific information, not a template. Describe the structure you're working with instead, like 'a trust with three unequal beneficiaries and a spendthrift clause,' and keep real names and numbers out of it.
Which tool: Estate planning practices tend to run leaner than litigation firms, so a paralegal's AI access is often whatever free ChatGPT or Gemini they set up themselves rather than a firm license — worth confirming with your supervising attorney what's actually approved before you rely on it daily.
Plain-language summary
Summarize a policy excerpt in plain language without adding legal advice.
Copy this prompt
I work in estate planning paralegal work. Help me create one plain-language summary that I can review and edit before use.
Relevant context:
- Role: Estate Planning Paralegal
- Work context: You prepare first drafts of will and trust documents from attorney notes, track which pages still need a client's signature and witnesses, and explain a beneficiary designation mismatch to a client who thought their estate plan was already settled.
Task:
Create a plain-language summary that separates what the policy text literally says from any interpretation of what it means.
Source material:
"Marcus from Support pinged me again about the AI-tools clause -- his team keeps asking what it actually means and getting a different answer from whoever they ask. Here's the piece everyone's stuck on: 'Employees may use approved communication tools for business purposes. Confidential information must not be shared with external systems unless authorized by policy and management.' Can you turn that into something his team can read in two minutes flat, split into what it literally says versus what people keep assuming it means? Last quarter somebody told his team it meant no ChatGPT ever, full stop, and that wasn't right, so I don't want another guess dressed up as an answer. Need it before his 9am standup Thursday."
Return only the finished work:
plain-language summary
Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent
Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.
Policy plain-language summary
Summarize an internal policy into plain do and don't points — no legal advice.
Copy this prompt
I work in estate planning paralegal work. Help me create one plain-language do points that I can review and edit before use.
Relevant context:
- Role: Estate Planning Paralegal
- Work context: You prepare first drafts of will and trust documents from attorney notes, track which pages still need a client's signature and witnesses, and explain a beneficiary designation mismatch to a client who thought their estate plan was already settled.
Task:
Turn a policy excerpt into plain-language do/don't points and a short 'ask a person if' list. Do not give legal advice.
Source material:
"Hey -- new AI-use policy just dropped from Legal and half of engineering already has questions in the Slack channel. Section 4.2 is the part that matters: 'Employees may use approved AI tools for non-confidential drafting work; no confidential, regulated, or client data may be entered into any AI tool absent written authorization from the Compliance Office; where classification is uncertain, the requestor shall escalate to the designated policy owner prior to use.' Renata in HR already asked if she can use it to draft an offer letter that references a candidate's visa status, and Owen on the vendor team wants to paste a supplier's SOC 2 report into ChatGPT to summarize it. I need something the whole team can actually read before Tuesday's town hall."
Return only the finished work:
1. plain-language do points
2. don't points
3. 'ask first if' list
Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent
Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.
Question list builder
Generate clarifying questions for a legal or compliance review request.
Copy this prompt
I work in estate planning paralegal work. Help me create one clarifying question list that I can review and edit before use.
Relevant context:
- Role: Estate Planning Paralegal
- Work context: You prepare first drafts of will and trust documents from attorney notes, track which pages still need a client's signature and witnesses, and explain a beneficiary designation mismatch to a client who thought their estate plan was already settled.
Task:
Build a review-intake question list for this addendum request without answering whether Section 4.2 is legally acceptable.
Source material:
"Hey — can you take a look at this before Friday? Corrigan Fulfillment sent over an updated data-sharing addendum for the Warehouse Ops contract, matter #LC-2291. Section 4.2 says they can share our inventory data with a subcontractor we've never worked with — no name, just 'an affiliate.' Procurement wants to sign this week so the shipping integration doesn't slip. Can you just tell me if Section 4.2 is okay to sign as-is?" — Dana, Procurement
Return only the finished work:
clarifying question list
Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent
Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.
How to use ChatGPT or Copilot for estate planning paralegals work
Start from your own rough version. Type or dictate what you would have written anyway, however messy. A prompt built on your real note beats a blank-page request every time.
Say who it is for and how long it should be. "Two sentences for a supervisor" and "a checklist a new hire can follow" produce completely different output from the same facts.
Strip the identifying details first. A client's actual asset list, account numbers, or family details from an estate planning questionnaire — that's privileged, client-specific information, not a template. Describe the structure you're working with instead, like 'a trust with three unequal beneficiaries and a spendthrift clause,' and keep real names and numbers out of it.
Read every specific it produces. AI fills gaps with plausible invention — dates, amounts, names, policy numbers. If you did not supply it, do not trust it.
Keep the ones that work. The value compounds when you stop rewriting the same prompt. Save it with a note about when to use it.
Practice, don't just copy
Get better at these — free
Copying a prompt is a start. AI Lunchroom walks you through the task step by step with your own work, checks what your AI tool gives back, and saves the prompts you will reuse. Free to start, no credit card.
is it okay to upload a client's will to ChatGPT to check the formatting
No — a signed or draft will is a real legal document with a real client's name, assets, and family structure in it, and once it's in a general AI tool you've lost control of where that copy lives. If you need formatting help, build a blank template with placeholder names and dollar amounts, then drop the client's real language in yourself afterward.
how much should I trust an AI-drafted trust provision
Not much on its own — AI can produce standard-sounding trust language that is subtly wrong for your state's rules on spendthrift clauses or trustee powers. Use it to get a rough structure moving faster, then have the attorney check every substantive clause against your jurisdiction's requirements before it goes anywhere near a client.
