Free CITE Prompt Builder · No signup
Choose the work product, record type, and practice context. Get a copy-ready prompt with explicit scope, output format, and pinpoint-citation rules—assembled in your browser.
Runs in your browser. Nothing is uploaded, and your matter details stay on this device unless you choose to run optional AI review. How your data is handled
Context
You are a litigation support analyst assisting insurance staff counsel in a first-party property (Hurricane claim) matter. …
Evidence
For every finding, cite page and line in the form `page:line` (e.g., 47:12–48:3) and quote the operative testimony verbatim. …
Free prompt builder
Choose the work, describe the record, then copy the finished prompt into the AI tool your firm has approved for the material.
Already have a prompt? Check its structure locally →Eligible prompts assemble with your practice priorities already composed in.
Start with the deliverable
Each prompt defines the assignment, output shape, and evidence standard before the record is reviewed.
Turn a complete deposition transcript into a topic-based page-line summary with precise cites, verbatim excerpts, admissions, exhibits, and unanswered questions.
Pull every admission, damaging concession, and 'I don't know' on your key issues out of a transcript — with verbatim Q&A and page:line cites.
Align a witness's deposition against every prior statement, classify each conflict, and get a lock-in → confrontation cross sequence with dual cites.
Turn the verified case record into a topic-based deposition outline with fact-locking goals, record-cited exhibits, conflict probes, and open-to-closed question sequences.
The review standard
CITE defines who the work is for, what the task is, the output shape, and the source standard. In today’s tools it creates copy-ready prompt text. In the planned workbench it becomes the workflow contract.
C — Context
Who you are, the matter, the document, the case posture.
I — Instructions
The precise task with explicit scope.
T — Template
The exact output format the AI must fill.
E — Evidence
Citation requirements — page:line, timestamp, Bates — plus the anti-hallucination directive: if it’s not in the record, say “not found in the record” — never infer.
Practice context
Your practice changes what a prompt prioritizes—not the requirement to show its source.
View all 12 practices →Privacy, source & review
A pinpoint makes a finding inspectable. It does not make the finding accurate.
Any Workbench pilot would require an explicit draft state and human approval before export or downstream use.
There is no file upload. Matter details are never put in links. Prompt text leaves only when you choose to run optional AI review.Read the full data path →
I’m Andrew. I’ve spent my career in Florida litigation support—depositions, transcripts, and the certified record.
CasePrompts started because “summarize this” is not a litigation workflow. The CITE library made the assignment clearer. The Workbench is the next step: pair that assignment with the record and make source review part of the product.
The public builder remains a practical, local-first tool, like depoaudio.com. It generates prompt text in your browser and does not accept case files. The private Workbench shown here is still in development and is not accepting matter files. Provider routing, retention, deletion, access controls, and legal terms must be set and disclosed before processing opens. See how today’s tools handle data.
More about me at andrewmayes.com or on LinkedIn. Something broken, or a template you wish existed? support@caseprompts.com
The Workbench would keep the page and line beside every finding, with human approval before anything leaves draft. It does not process files today.