How to Write Effective Prompts for Eve

Overview

Most of what gets described as "Eve isn't giving me good answers" is a prompting problem rather than a product problem. The gap between a mediocre Eve response and an excellent one is almost always something you control: what you asked for, what you gave Eve to work with, and whether you said what to do when the answer isn't there.

This guide covers both halves of that. Part 1 is the mental model for how Eve reads what you send. Everything after it turns that model into concrete techniques with Eve-specific examples.

This applies anywhere Eve takes a written instruction: Talk to Eve, chat agents, drafting agents, and auto agents.

Part 1: How Eve reads your prompt

Eve is a brilliant new colleague with zero case history

The single most useful frame: Eve is a really smart, quick paralegal who started at your firm this morning. Eve has read every document you've put in front of it and nothing else. Eve doesn't know your client, your opposing counsel, your judge's preferences, your firm's shorthand, or what got decided in last Tuesday's case strategy meeting.

That's why "check that email about the incident" fails and "review the email from Jane Smith dated March 15, 2023, titled 'Incident Report: Warehouse'" succeeds. Eve isn't being difficult. Eve has forty-two emails and no way to know which one you mean.

The colleague test: before sending a prompt, ask whether a smart new hire with no case background could execute it correctly. If they'd have to come back with a question, the prompt needs that answer in it.

Context is a budget, not a bucket

There's a tempting assumption that giving Eve more documents produces better answers. Past a point, it doesn't. Every document you make available is another place Eve has to look, and irrelevant material competes for attention with the material that matters.

Treat your context selection as an assignment, not an archive. Hand a paralegal a single deposition and ask for a summary, and you'll get a tight one. Hand them a banker's box and ask them to summarize the deposition inside it, and you'll wait longer for something less focused.

The practical rule: match context scope to question scope.

Eve is agreeable by default

Ask Eve to find something and Eve leans toward finding it. Phrase a request as "find all instances of HR complaints" and you've quietly told Eve that instances exist and the job is to produce them. If the documents contain none, what comes back is a marginal match carrying more confidence than it has earned.

The fix is one clause long, and it's technique 4 below. The mechanism is worth understanding, because it explains a whole category of Eve errors that otherwise look random.

Eve interprets literally, and format is an instruction

Eve does what you say, not what you meant. That cuts both ways: sloppy phrasing produces sloppy output, and precise phrasing produces precise output. "Summarize the medical records" invites Eve to decide what a summary is. "Summarize the medical records in under 200 words, focused only on treatment received after the incident date" leaves nothing to interpretation.

The same goes for structure. Don't specify a format and Eve picks one, which may not be the one it picked last time you asked a similar question. Format instructions aren't cosmetic. They're how you get consistency across a caseload.

Part 2: The anatomy of a strong prompt

Not every prompt needs every part. Simple lookups don't need scaffolding. But when a request matters, a billing breakdown, a claims analysis, anything you'll rely on, these five components are the checklist.

1. Task. The specific, concrete thing you want Eve to do. Lead with it. One deliverable, stated plainly, so Eve knows what it's being asked before reading anything else.

2. Objective. Why you need it and what it's for. This is the piece people skip and the one that does the most work. Eve uses the purpose behind a request to make better judgment calls on the hundred small decisions you didn't specify. "Summarize this deposition" and "summarize this deposition because I'm preparing to cross-examine him" produce very different summaries.

3. Context. The case background Eve needs and can't get from the documents alone. Case type, procedural posture, what's already established, who the parties are, what matters strategically.

4. Format. How you want the answer returned. Table, bullets by category, narrative prose, a specific set of fields per entry.

5. Additional instructions. The edge cases, exceptions, and special handling. Everything that starts with "and if…": how to handle ambiguity, what to do when something isn't found, what to exclude, what to flag rather than assume.

The template

[Task] Do X. [Objective] I need this for Y. [Context] Here's what you should know about the case. [Format] Return it like this. [Additional instructions] Handle these situations this way.

The labels are only here to show the parts. You don't need them in a real prompt; Eve reads it fine as ordinary paragraphs.

Worked example

Weak:

Find the discrimination evidence.

Strong:

Identify every instance in these documents where the plaintiff reported discriminatory treatment to management or HR [task].

I'm building the factual record for a summary judgment opposition, so I need this to be exhaustive and defensible: every instance I can point to, with a citation I can verify [objective].

This is a wrongful termination case with an employment discrimination claim. The HR records span 2019 to 2023. The plaintiff alleges she reported discriminatory treatment repeatedly before being terminated in November 2023 [context].

Give me a table with one row per instance: date, who she reported to, the medium (email, meeting, formal complaint), a one-sentence description, and the source document with page number [format].

If an instance is ambiguous, where it's unclear whether she was describing discrimination or a general workplace complaint, put it in a separate section below the main table rather than mixing it in. If there are no instances at all, tell me that directly instead of stretching to fill the table [additional instructions].

The second prompt is longer, and that's the point. Ninety seconds of writing saves two rounds of correction plus a false positive you might not have caught.

Why the order matters

Task first, so Eve knows what it's doing before it processes why or under what constraints. It's the same reason you'd tell a paralegal "pull every discrimination complaint" before launching into the background. Objective second, because the why colors how Eve reads everything after it. Context, format, and instructions then narrow the work from broad to specific.

If you've built a drafting agent, this structure will feel familiar. Drafting agents already split the work into what you're producing, what documents inform it, and an instructions field for the specific tweaks. This is the same shape, written out longhand in chat.

Part 3: Core techniques

1. Be specific about documents, dates, parties, and facts

Name things. Use exact file names, full party names, specific dates, and page ranges when you have them.

• Instead of "what does the evidence say?" → "What does John Smith's deposition transcript say about the termination meeting on March 15th?"

• Instead of "check the employment agreement" → "Looking at pages 15 to 20 of the employment agreement, what are the employee's non-compete obligations?"

• Instead of "the medical stuff from last year" → "Dr. Ramirez's treatment notes from January through June 2024"

You'll know it's working when Eve's citations point to exactly the documents you named.

2. Explain the why, not just the what

The most underused technique here and one of the highest-leverage. Telling Eve the purpose behind a request lets it make sensible decisions in the gaps.

• "Summarize this deposition" produces a generic summary.

• "Summarize this deposition. I'm preparing to cross-examine this witness and I need to know where his account contradicts the police report" produces a summary organized around contradictions, with the details that matter for impeachment surfaced first.

Same document, same nominal task, completely different usefulness.

3. Scope your context deliberately

Before sending a prompt that matters, open the Select context menu and check whether the file list matches the question. Deselect what isn't relevant.

This is the best effort-to-payoff ratio in the guide, and it's the one people skip because it happens outside the text box. Narrowing context reduces noise, sharpens citations, and often speeds up the response.

One caveat worth internalizing: narrowing context is not the same as narrowing the question. Deselect the police report, then ask about liability, and Eve will answer confidently from what's left. Eve is working from the assignment you gave it, so make sure the assignment includes what the question actually requires.

4. Give Eve an out

Add an explicit negative-result clause to any request that asks Eve to find something.

• "Find all instances of HR complaints in these documents. If there are none, tell me there aren't any."

• "Identify any prior accidents referenced in the records. If the records don't mention any, say so rather than inferring from context."

• "List the treatment gaps longer than 30 days. If treatment was continuous, state that."

Success looks like Eve plainly saying "I did not find any," which is a real answer and often a valuable one.

A close cousin: give Eve permission to flag uncertainty. "If you're not confident about a date because the record is illegible, mark it as uncertain instead of picking the most likely reading."

5. Specify your output format

Tell Eve exactly how you want information structured. It makes responses easier to review, and it makes results comparable across matters when you reuse the prompt.

Respond with a section header for each type of claim, and under each, a bulleted list of the potential claims of that type with a one-sentence description of each. End with a short paragraph on which claim you'd rank strongest and why.

Two refinements that can help:

Ask for what you want, not what you don't. "Write this in flowing prose paragraphs" works better than "don't use bullet points." A negative instruction makes Eve hold the forbidden thing in mind; a positive one just describes the target.

Your prompt's style influences the output's style. Write a terse, clipped prompt and you'll tend to get terse output. Write a formal one and you'll get formal output. If you want a particular register, model it.

6. Show Eve an example

If you have a format you want matched, paste one in. A single worked example does more to control structure than three sentences describing that structure, and two or three examples covering different shapes do better still.

Here's how I formatted the visit chronology for our last case:

3/14/2024 - Dr. Patel, Northside Orthopedics - Initial consult; complaints of lower back pain radiating to left leg; ordered MRI. [Bates 00412]

Build the chronology for this matter's records in exactly that format.

This is also how you bring firm conventions to Eve without writing a style guide.

7. Break complex work into chunks

If a request contains the word "and" three times, it's probably three prompts.

Instead of: "Review all the depositions, create a timeline, identify contradictions, and draft a witness list."

Do this:

1. "Create a chronological timeline of events based on all deposition testimony."

2. Review the timeline, then: "Now identify where the witnesses' accounts contradict each other on these events."

3. Review the contradictions, then: "Draft a witness list ranked by how central each witness is to the disputed facts."

Two things happen. Each response is better, because Eve puts full attention on one deliverable. And you catch errors at step one instead of discovering at step three that the whole chain was built on a bad timeline.

Draft, review, refine, as separate prompts, beats asking for a polished final product in one shot.

8. Ask Eve how to ask

If you're not sure how to frame something, ask directly.

• "I'm trying to extract a timeline of key events from these medical records. How should I ask you for that?"

• "What's the best way to request a case strategy memo from you? What information would you need from me?"

Eve will usually come back with a prompt template you can edit and reuse. It's the fastest way to get to good prompts without studying prompting.

9. Verify, always

Eve supports your judgment rather than replacing it. Verification isn't a trust issue, it's the job.

Use the citations. Eve's responses carry citation badges linking back to the exact source passage. Hover one for the source details, click it to open the source. This is faster than re-reading the document and it's the single best habit to build.

Ask Eve to check its own work. These prompts genuinely catch errors:

• "Can you check your work?"

• "Provide direct quotes from the documents to support each of these claims."

• "Review your timeline and confirm each date appears in the source materials."

• "Explain how you reached that conclusion."

Push back when something's off. If a response is wrong or incomplete, say specifically what's wrong. "That's not right" gives Eve nothing. "The third entry cites the wrong provider, that visit was with Dr. Chen, not Dr. Ramirez. Re-check the provider on each entry against the letterhead in the source records" gives Eve something to act on.

10. A note on "think step by step"

Older prompting guidance told you to add "think through your reasoning step by step before answering." That advice is largely obsolete. Eve works through multi-step problems without being told to, and the instruction rarely improves quality now.

It's still useful for a different reason: auditability. When you need to see how Eve reached a conclusion, because you're going to rely on it or because you suspect it's wrong, asking Eve to show its reasoning gives you something to check. Use it when you need transparency, not as a default quality booster.

Part 4: Prompting when Research is on

Eve's Research feature changes how you should prompt, because Eve is no longer limited to your matter documents. Research covers two capabilities, and both are toggled separately in chat:

Trusted Web Search pulls current public information: regulatory updates, government benefit rates, medical literature, employer background, all with URL, title, publish date, and excerpt on every result. With the Web Search toggle on, a source selector lets you pick a collection before you ask.

Legal Research brings case law and court opinion research into the same chat. The chat control is labeled Case law search, in the Research tab. Citations come back in Bluebook form, and clicking one opens the opinion in a panel inside Eve with its treatment badge, so you can confirm the authority without leaving. Legal Research is generally available.

Both require a firm admin to enable them in Settings → Research.

Three prompting adjustments:

Say where the answer should come from. When documents and web sources are both in play, be explicit about which you want. "Based only on the medical records in this matter, what treatment did the plaintiff receive?" is a different question from "What is the standard treatment protocol for this condition?" Mixing them without saying so gets you a blended answer that's hard to verify.

Pick the collection that fits the task. Statutes & Regulations for regulatory lookups, Medical Research for clinical questions, Academic Literature for expert commentary and scientific evidence, Business Filings for entity research, All Web for broad research. Your firm admin may have built custom collections too, and they appear in the same selector. Choosing a collection is itself a form of prompting: it narrows the source universe before Eve reads a word, exactly like narrowing your context selection does for documents.

Set the jurisdiction on case law questions. Case law search runs in Smart Mode by default, taking its scope from the matter, and you can refine it in plain language: "also include the 10th Circuit," "go back to 2015," "switch to Delaware." You can also set jurisdiction and date range manually.

Verify web and case law citations the same way you verify document citations. Every web result carries its source details. Every case law citation opens inside Eve so you can read the opinion and check the treatment badge. Use them.

Part 5: Prompt to Eve's strengths

Structuring requests around what Eve does well is itself a prompting technique.

Eve is strong at document analysis and information extraction, pattern recognition across large record sets, chronology and timeline construction, legal reasoning grounded in documents you've provided, drafting in a learned style, and synthesizing case law and public sources when Research is enabled.

Know the limits:

Precise math. Eve is unreliable at complex calculations. Have Eve extract and organize the numbers, and verify the arithmetic yourself.

Images without text. Eve reads text out of images, including photographs of documents, screenshots, and scanned records. Eve cannot describe the visual content of a photo. A picture of a totaled vehicle will never produce a meaningful description of the damage. If the information exists only as an image and not as text, it isn't available to Eve.

Anything outside its context. Eve answers from your selected documents plus, when enabled, the web and case law sources you've allowed. Eve doesn't know what's in your case management system, your inbox, or your head unless you've said so.

Part 6: Troubleshooting

Quick reference

Before you hit send on anything that matters:

• Would a smart new hire with no case background understand this?

• Have I named specific documents, dates, and parties?

• Did I say why I need this, not just what I want?

• Is my context selection matched to the question?

• Is this one deliverable, or three?

• Did I specify the output format?

• Did I give Eve an out?

• After the answer: did I click the citations?