---
title: "Example Prompts"
description: "Example prompts for researching decisions, judges, parties, patents, and new filings with the Docket Navigator MCP."
sidebar:
  badge:
    text: "Coming Soon"
    variant: note
---

import { Steps } from '@astrojs/starlight/components';
import PromptCard from '../../../../components/PromptCard.astro';
import PromptCardList from '../../../../components/PromptCardList.astro';

Connecting Docket Navigator to your organization's AI provider platform lets you describe a research assignment in ordinary language and get an answer built from Docket Navigator's structured litigation data, with citations and links back to the underlying cases, orders, and profiles.

The Docket Navigator MCP connects Docket Navigator data and research tools to your organization's AI provider platform. The AI platform interprets your request, decides how to use those tools, and generates the response you see. Results may vary depending on the AI provider and model your organization uses.

:::note
Review important results for factual, doctrinal, and procedural relevance, and use Docket Navigator citations and links to verify the underlying source material.
:::

:::tip
Two habits make everything below work better. Ask why a result is included when you want to see the reasoning. Ask where a number comes from when the answer contains a statistic, so you can confirm the population and time period being counted.
:::

Some of these work well hands-free, in the car or out for a run. Ask for the short version and say you are listening rather than reading. Spoken answers are for orientation. When something matters, ask the platform to flag it so you can open the Docket Navigator source later.

## Find the decisions you need

This is the same kind of research described in [Fact-Pattern Decision Research](/mcp/fact-pattern-research/).

<PromptCardList>
<PromptCard title="Find decisions by the motion type and the legal issue together" icon="scale">
	<span slot="try">Find decisions on motions to dismiss for failure to state a claim that address patent eligibility under Section 101.</span>

	Typically a set of coded orders with case names, filing dates, and links into Docket Navigator. Docket Navigator codes the procedural vehicle and the legal issue separately, so the platform can hold both conditions at once rather than relying on wording in the document. Counts on this kind of search are counts of coded document events, so the same order can appear once for each motion type coded to it.

	<span slot="next">Limit those to the last three years and tell me which way they came out.</span>
</PromptCard>

<PromptCard title="Narrow a decision set to the judge you are appearing before" icon="user-round">
	<span slot="try">Of those Section 101 dismissal decisions, which ones were issued by Judge Connolly in Delaware?</span>

	Typically a short list of that judge's orders on the same motion type and issue, with case names, dates, and links. The judge, motion type, and legal issue conditions combine, so the result is the intersection rather than a broader list you have to filter by hand.

	<span slot="next">Summarize what happened in each one and note any that came out differently.</span>
</PromptCard>

<PromptCard title="Read what the order actually says" icon="file-text">
	<span slot="try">Open the order in that first case and show me the part about the eligibility analysis.</span>

	Typically bounded text drawn from the filing itself, along with a link to the document in Docket Navigator. This works when Docket Navigator holds the filing and your subscription allows you to open it. Image-only pages may return no text. Use it to confirm that a description of a decision matches the document.

	<span slot="next">Does that reasoning appear in any of the other decisions on this list?</span>
</PromptCard>

<PromptCard title="Check why a result is in your set" icon="circle-help">
	<span slot="try">Why is that third case in these results? I would not have expected it.</span>

	Typically an explanation in terms of the Docket Navigator coding that matched, such as the document type recorded for the order and the legal issue annotated to it, with a link so you can inspect the record yourself. This is a good way to spot a case that matched on a technicality.

	<span slot="next">Take that one out and show me what the set looks like without it.</span>
</PromptCard>

<PromptCard title="Combine two legal issues, either way round" icon="combine">
	<span slot="try">Find decisions that address both Section 101 eligibility and indefiniteness in the same order.</span>

	Typically a much smaller set than either issue alone, because the conditions combine. You can also ask for decisions addressing either issue, which returns a far larger set. Say which one you want, since the two questions produce very different numbers from the same two issues.

	<span slot="next">Now show me the ones addressing either issue rather than both.</span>
</PromptCard>

<PromptCard title="Find the rulings on expert testimony in a party's cases" icon="microscope">
	<span slot="try">Find the Daubert rulings on expert testimony in cases involving Apple.</span>

	Typically the coded orders on motions to exclude expert testimony in that party's cases, with case names, dates, and links. Whether an expert was excluded is recorded in the ruling, so ask for the outcome rather than assuming the motion succeeded. Expert names appear in the text of the orders rather than as a searchable field, so ask the platform to open an order when you need the name.

	<span slot="next">Open the most recent one and tell me what the court decided.</span>
</PromptCard>

<PromptCard title="Find the remedies awarded, not just the rulings" icon="banknote">
	<span slot="try">What damages awards has Docket Navigator recorded in patent cases, and can you show me recent examples?</span>

	Typically a set of recorded remedies with the case citation for each. Remedies are recorded separately from determinations and case outcomes, so a damages award is its own record with its own type. Recorded awards reflect what was awarded at the time; later reconsideration or reversal on appeal is not reflected in the recorded amount.

	<span slot="next">Show me the injunctions and ongoing royalties instead.</span>
</PromptCard>

<PromptCard title="See what was actually awarded against a company" icon="building-2">
	<span slot="try">What damages have been awarded in patent cases involving Samsung, and how large were they?</span>

	Typically the recorded awards with the dollar amount and the case citation for each. Awards are recorded as entered at the time, so a figure later reduced, vacated, or reversed on appeal still appears at its original amount. Check the case before quoting a number.

	<span slot="next">Which of those were enhanced damages or attorney fee awards?</span>
</PromptCard>

<PromptCard title="Check a firm's record in Board proceedings" icon="briefcase">
	<span slot="try">How many IPR institution decisions came out of proceedings where Sidley Austin appeared?</span>

	Typically the institution decisions in that firm's proceedings with case names and decision dates. These are counts of coded decisions rather than of unique proceedings, so ask the platform to separate instituted from denied outcomes and to state the date range before treating any figure as an institution rate.

	<span slot="next">Split those into instituted and denied, and give me the last three years only.</span>
</PromptCard>
</PromptCardList>

## Prepare for a hearing, a motion, or a deadline

This is the same kind of research described in [Pre-Hearing Decision Research](/mcp/pre-hearing-decision-research/).

<PromptCardList>
<PromptCard title="See how your judge has handled the motion you are about to argue" icon="gavel">
	<span slot="try">I have a motion to stay pending IPR in front of Judge Gilstrap next month. How has he ruled on those?</span>

	Typically the judge's decisions on that motion type with outcomes and dates, plus links to the orders. Ask for the time period and the number of decisions behind any rate, since a grant rate over five years and a grant rate over twelve months can differ substantially.

	<span slot="next">How does that compare with the rest of the Eastern District of Texas?</span>
</PromptCard>

<PromptCard title="Compare your judge with the surrounding court" icon="git-compare">
	<span slot="try">Compare Judge Gilstrap's outcomes on stay motions with the rest of his district and with Delaware.</span>

	Typically an outcome breakdown for each population, often with a chart. Ask the platform to state the denominator and the date range for each group, and to say how the comparison population was defined, since a judge, the remainder of a district, and a separate district are three different denominators.

	<span slot="next">Show me the decisions behind the Delaware column.</span>
</PromptCard>

<PromptCard title="Find out how long this court takes to get to a milestone" icon="timer">
	<span slot="try">In the District of Massachusetts, how long does it typically take a patent case to reach claim construction, and how long to trial?</span>

	Typically a distribution for each litigation milestone, including the number of occurrences, the median, the quartiles, and the average, measured in months. The occurrence count is the denominator, and these figures cover cases filed from a stated start date onward, so ask for both before relying on the numbers.

	<span slot="next">How does the median time to claim construction compare with the judge assigned to my case?</span>
</PromptCard>

<PromptCard title="Check where a parallel PTAB proceeding stood on a given date" icon="git-branch">
	<span slot="try">When the court ruled on the stay motion in this case, had the PTAB instituted review on the asserted patents?</span>

	Typically the posture of each asserted patent as of the order date, reported as instituted, institution denied, still pending, or no petition on file, with the PTAB record behind it. Posture is resolved for each patent separately, so a case with several asserted patents can be mixed. Ask for the posture as of the order date rather than as of today.

	<span slot="next">Which of those patents had a petition on file but no decision yet?</span>
</PromptCard>
</PromptCardList>

## Understand a judge, a court, or a filing trend

This is the same kind of research described in [Judge and Court Intelligence](/mcp/judge-and-court-intelligence/).

<PromptCardList>
<PromptCard title="Ask what a judge actually does in patent cases" icon="landmark">
	<span slot="try">Give me a picture of Judge Connolly's patent litigation practice: what motions he decides most, and how they come out.</span>

	Typically a profile-level summary drawn from Docket Navigator's coded orders, with links into the underlying decisions. For any rate, ask what was counted and over what period. Docket Navigator distinguishes motion outcomes from case outcomes, so a judge who denies most of a motion type is not necessarily a judge whose cases resolve a particular way.

	<span slot="next">Show me his ten most recent claim construction orders.</span>
</PromptCard>

<PromptCard title="Compare outcome definitions before relying on a number" icon="equal">
	<span slot="try">What is the difference between the case outcomes and the accusation outcomes you just showed me?</span>

	Typically an explanation that an accusation is a single patentee, patent challenger, and patent combination, and that case outcomes are assembled from the accusation outcomes once they are resolved. Outcomes shown are the original lower court results, and post-appeal outcomes are not currently counted. This distinction changes what a win rate means, so it is worth settling early.

	<span slot="next">Recalculate that using accusation outcomes and tell me if the picture changes.</span>
</PromptCard>

<PromptCard title="Track filing volume and where cases are being brought" icon="chart-column">
	<span slot="try">How many patent cases has this plaintiff filed each year, and in which districts?</span>

	Typically a count by year and by court, with the cases behind it available as a list with citations. Ask which date basis is being used, since the case filing date and the termination date produce different pictures of the same campaign.

	<span slot="next">Show me the cases from the two busiest years.</span>
</PromptCard>
</PromptCardList>

## Build the picture around a case, a patent, or a party

Related walkthroughs include [Party and Attorney Track Record](/mcp/party-and-attorney-track-record/) and [Case Pitch Builder](/mcp/case-pitch-builder/).

<PromptCardList>
<PromptCard title="Get oriented on a case you just heard about" icon="folder-open">
	<span slot="try">Tell me about Approov Ltd. v. Guardsquare NV in the District of Massachusetts.</span>

	Typically the parties, the assigned judge, the asserted patents, the cause and nature of suit, and the filing date, with a link to the case in Docket Navigator and access to its docket entries, pleadings and orders, determinations, claim constructions, remedies, and accusation outcomes.

	<span slot="next">What has been filed in it so far, and what is the patent it asserts?</span>
</PromptCard>

<PromptCard title="Catch up on a case while you are away from your desk" icon="headphones">
	<span slot="try">I'm out for a run and I've got a call on the Katana Silicon case this afternoon. Catch me up on what's happened in the last couple of months. Keep it short, I'm just listening.</span>

	Typically a brief spoken summary of recent activity: what was filed, what the court decided, and what is still outstanding. Telling the platform you are listening rather than reading helps, since the same request at a desk can return a long list with links. Ask for highlights rather than everything, because an active case can carry dozens of coded entries. You can ask for dates to be read out, and you can ask the platform to note the orders worth reading so you can open the Docket Navigator sources when you are back at a screen.

	<span slot="next">Just the orders the judge actually issued, and tell me which one I should read before the call.</span>
</PromptCard>

<PromptCard title="Trace what has happened to a patent across proceedings" icon="route">
	<span slot="try">What has happened to this patent in litigation? Any determinations, claim constructions, or PTAB challenges?</span>

	Typically the recorded determinations on infringement, validity, and enforceability, any construed terms with their constructions, and PTAB activity, each with a link to the source. Determinations are recorded by patent number rather than by individual claim, and a patent can have more than one determination, including ones that were later revisited.

	<span slot="next">Which terms were construed, and in which orders?</span>
</PromptCard>

<PromptCard title="Build an opponent's track record before you respond" icon="swords">
	<span slot="try">We received a demand letter from Taasera Licensing. What is their litigation history, and how have the patents they assert held up?</span>

	Typically their cases across courts with outcomes, the determinations made on their patents with dates, and a source link for each finding. This picture is built from what courts decided. It does not include the content of motion papers or attribute arguments to individual attorneys.

	<span slot="next">Which of their patents survived that invalidity order?</span>
</PromptCard>

<PromptCard title="Assemble the litigation background for a pitch" icon="presentation">
	<span slot="try">A new case was just filed against a client of ours. Put together the background I would need for a pitch: the asserted patents' litigation history, the judge, and the other side's record.</span>

	Typically a structured summary combining the case details, the history of each asserted patent, the assigned judge's relevant record, and the opposing party's prior cases, with a citation or link for each factual claim. Because this material is client facing, check the individual sources rather than the summary alone.

	<span slot="next">Which three of those prior decisions are most worth reading before the meeting?</span>
</PromptCard>

<PromptCard title="Follow a law firm or an attorney across cases" icon="users">
	<span slot="try">Which patent cases has this firm handled for patentees in the last five years, and how did they come out?</span>

	Typically the firm's cases in that role with outcomes and links. Docket Navigator records a separate outcome for each role, so a firm that has represented both patentees and patent challengers will have a separate record for each side rather than one blended figure.

	<span slot="next">Split that by district and show me where they file most.</span>
</PromptCard>

<PromptCard title="Find cases that reached a particular stage" icon="flag">
	<span slot="try">Show me patent cases in this district that actually reached a jury trial.</span>

	Typically cases narrowed by the coded litigation milestone rather than by guesswork about the docket. Milestones are defined by the filing of particular documents, so a jury trial milestone indicates that trial-stage documents were filed rather than giving the exact trial date.

	<span slot="next">Of those, which ones produced a damages award?</span>
</PromptCard>
</PromptCardList>

## Watch what is new and act on it

This is the same kind of research described in [Docket Report Monitoring](/mcp/docket-report-monitoring/).

<PromptCardList>
<PromptCard title="Triage today's Docket Report against a watch list" icon="file-spreadsheet">
	<span slot="try">There's a spreadsheet in this folder, watchlist.xlsx, with the clients, opposing parties, patents, judges and firms my group tracks. Go through today's Docket Navigator patent Docket Report and tell me which of today's activity touches anything on that list. I need the whole report covered, not a sample of it.</span>

	Typically a brief naming only the report items that match, with what happened, which watch-list row caused the match, why that row is on the list, and a link back to Docket Navigator. Coverage should be the full current report rather than a sample. Some matches are not visible from the caption alone, so the platform may need the underlying case record to confirm a tracked party, judge, or counsel.

	<span slot="next">Where did that case come from? We're not in it and I don't recognize the parties.</span>
</PromptCard>

<PromptCard title="Read today's activity against your own watch list" icon="eye">
	<span slot="try">Go through yesterday's Docket Navigator activity and tell me if anything touches these clients or these patents.</span>

	Typically a short answer naming only the filings that match, or a clear statement that nothing did. New case entries usually include the parties, the asserted patents, the assigned judge, the court, and an editorial summary of the complaint, with links to the filing.

	<span slot="next">Take the second one and pull the litigation history of the patent it asserts.</span>
</PromptCard>

<PromptCard title="Screen new filings for a specific kind of case" icon="scan-search">
	<span slot="try">In the last few days of new patent filings, were there any ANDA or biosimilar cases?</span>

	Typically the matching new filings with the nature of the claim identified, the asserted patents, and the parties. Coverage reflects what has been processed into Docket Navigator, so very recent filings can appear before every field has been coded.

	<span slot="next">Give me the patents asserted in the ANDA case and who else has litigated them.</span>
</PromptCard>

<PromptCard title="Save the research so you can come back to it" icon="bookmark">
	<span slot="try">Save this research to a Docket Navigator binder so I can pick it up next week.</span>

	A binder created in Docket Navigator, with your searches saved as tabs and a link to open it. This changes Docket Navigator, so it happens when you ask for it. The conversation itself is not saved research. The binder is.

	<span slot="next">Add a tab with the judge's decisions on the same motion.</span>
</PromptCard>

<PromptCard title="Turn a saved binder into ongoing monitoring" icon="bell">
	<span slot="try">Set up a daily alert on that binder so I hear about new activity.</span>

	An alert created on the binder at the frequency you name, with confirmation of the setting. Changing an existing alert shows you the before and after first and applies the change only after you confirm.

	<span slot="next">Change it to weekly on Mondays instead.</span>
</PromptCard>
</PromptCardList>

## Work beyond patent litigation

<PromptCardList>
<PromptCard title="Follow a brand owner's trademark enforcement record" icon="tag">
	<span slot="try">How many trademark cases has Chanel, Inc. brought in federal court, and where does it file them?</span>

	Typically the case count with the cases themselves listed by citation, court, and filing date, and links into Docket Navigator. Party names are matched as recorded in the case, so ask the platform to confirm the exact party name it used when a company appears under several variants.

	<span slot="next">Show me the ones filed in the Southern District of Florida in the last two years.</span>
</PromptCard>

<PromptCard title="Research opposition and cancellation practice at the TTAB" icon="stamp">
	<span slot="try">What oppositions has Monster Energy filed at the Trademark Trial and Appeal Board, and how have they turned out?</span>

	Typically the Board proceedings with their proceeding numbers, filing dates, and links. Docket Navigator's trademark library covers Trademark Trial and Appeal Board proceedings alongside federal court cases, so oppositions, cancellations, and ex parte appeals can be searched the same way district court cases are.

	<span slot="next">Compare that with the district court cases the same company has filed.</span>
</PromptCard>

<PromptCard title="Track a copyright plaintiff or defendant across cases" icon="copyright">
	<span slot="try">What copyright cases has Sony Music Entertainment been involved in, and in which courts?</span>

	Typically the cases with citations, courts, and filing dates, and links into Docket Navigator. The same party, court, and date filters used in patent research apply here.

	<span slot="next">Narrow that to cases filed in the Southern District of New York.</span>
</PromptCard>

<PromptCard title="Survey trade secret, antitrust, or securities filing activity" icon="lock">
	<span slot="try">How many trade secret cases were filed in federal court last year, and which courts saw the most?</span>

	Typically a count for the period you name with the cases listed behind it. State the year or date range in your request, since the count depends entirely on the filing window being used. The same approach works for antitrust and securities litigation.

	<span slot="next">Show me the antitrust cases from the same period instead.</span>
</PromptCard>

<PromptCard title="Look at proceedings before the Unified Patent Court" icon="globe">
	<span slot="try">What has been filed at the Unified Patent Court since the start of 2024, and which divisions are busiest?</span>

	Typically proceedings from the UPC divisions with their case numbers, filing dates, and the division involved. The UPC record has its own structure and vocabulary, so ask the platform what it is counting before comparing UPC figures with United States district court figures.

	<span slot="next">Show me the proceedings in the Munich local division.</span>
</PromptCard>

<PromptCard title="Check the US family of a patent asserted at the UPC" icon="git-fork">
	<span slot="try">This patent is being asserted against us at the Unified Patent Court. What are its US family members, and have any of them been litigated here?</span>

	Typically the patent family from the European Patent Office record, with the US members identified, followed by whatever litigation record exists for those US patents. The family list mixes granted patents with published applications, so ask which are granted before relying on the list. A US family member with no results has not been asserted in a court, the ITC, or the PTAB, which is not the same as the patent not existing.

	<span slot="next">Take the granted US ones and tell me if any have been through the PTAB.</span>
</PromptCard>
</PromptCardList>

## Verify important results

Before you rely on any proposition from this workflow:

<Steps>

1. Open the cited Docket Navigator source.

2. Confirm the source supports the description you were given.

3. For any statistic, check the denominator, the date period, and the comparison population.

4. For anything going into work product or client-facing material, work from the underlying document rather than the summary.

</Steps>

:::tip[The key idea]
Describe the assignment you actually have, and let Docket Navigator's structured record supply the decisions, outcomes, and sources behind the answer.
:::