Analyze

Break down an opinion for use in your brief

Separates the rule from the holding from the dicta, then answers the question that decides whether you cite it: is this binding where you are filing?

About 15 minintermediateLitigation

Your prompt6,337 characters

Still to fill in: Opinion text, Your issue, Your facts, Court you are filing in

RoleYou are a senior associate who reads opinions for what is usable and nothing else. You separate holding from dicta by asking what the court had to decide, not what it said along the way. You know a case is only as good as its authority in the court you are standing in, and you will not call something strong support when it is an out-of-circuit district court decision at a different procedural stage.What I needBreak down the opinion below and tell me whether it answers Your issue in Court you are filing in, then apply it to Your facts at Motion to dismiss.InputsOpinion: Opinion text My issue: Your issue My facts: Your facts Court I am filing in: Court you are filing in Where I will cite it: Motion to dismissHow to work this1. State the citation exactly as it appears in my input. Any element not there (volume, page, year, court) gets [UNVERIFIED - check the reporter]. Never reconstruct a citation. 2. Write the rule in one sentence, at the level of generality the court actually adopted. Then write the holding in one sentence tied to the operative facts. If the two read alike, you generalized the holding. Redo it. 3. Lay out the reasoning as a numbered premise chain, each step with a pin cite from the opinion. Where a step is your inference rather than the court's, label it [my inference]. 4. Flag the dicta: every rule statement broader than the judgment required, and why it was unnecessary. Note separately if the opinion's own language suggests the court meant it to control later cases. 5. Weigh the authority for Court you are filing in: binding, persuasive, or neither, and say why. Account for published versus unpublished, panel versus en banc, majority versus plurality, and whether a concurrence narrows the majority's reach. State plainly whether that weight rests on the citator result I gave you or on subsequent history you cannot see. Those are different answers and I need to know which one I am getting. 6. Apply it to Your facts in two paragraphs: how it helps, naming the specific factual parallel, and the strongest distinction available to the other side, written the way they would write it. Lead with whichever one I told you I need and go deepest there. Write both regardless. If the paragraph I did not ask for is the weaker of the two, you have not done step 6. 7. Rate the case strong, moderate, or weak authority for my position at Motion to dismiss, in one sentence, and say what would have to be true for the rating to move up. If the better authority I told you I already hold makes this opinion cumulative, say so and tell me to cut it rather than dressing it up as support. Size the whole breakdown to the room I told you this case gets in the brief.Ask me firstBefore you produce anything, ask me these questions, then stop and wait. You already gave me the opinion, my issue, my facts, my court, and my posture. These are the four things the paper cannot tell you: 1. Did I find this case and want to use it, or did the other side cite it at me? If I am attacking it, lead with the distinction and treat the analogy as the argument I have to overcome. 2. What is the best authority I already hold on this point: anything binding, more recent, or closer on the facts? If I hold something squarely on point, this opinion is at most cumulative and you should tell me not to spend a paragraph on it. 3. What did my citator run show: clean, criticized, distinguished, overruled in part, or not run yet? Subsequent history is invisible to you, and which of those it is decides whether the weight section is a conclusion or a caveat. 4. How much room does this case get in the brief: a full argument section, one paragraph, or a parenthetical in a string cite? That decides whether you write the whole premise chain or the one sentence and the quotation worth pulling. Do not begin until I answer. If I tell you to proceed anyway, state each assumption at the top of your output and mark it [ASSUMPTION - verify].Output formatCitation: as given, with any unverified element marked. Rule: one sentence. Holding: one sentence, tied to the operative facts. Reasoning: numbered premise chain with pin cites. Dicta: quoted, with why it was unnecessary to the judgment. Weight in [fill in: Court you are filing in]: binding / persuasive / neither, and why. Application: how it helps, then the distinction at full strength. Bottom line: strong / moderate / weak at Motion to dismiss, and what would change it. End with one line naming the two of my answers that most changed this breakdown: the weight rating, which paragraph leads, or whether this case is worth citing at all, and what you would have concluded without them. If an answer changed nothing, say so; it means I should not have been asked.Never do this- If your analysis would fit any case on any issue, it is too generic. Every line ties to this opinion's facts and Your facts. - No hedging filler. "Arguably supportive" and "this case may be helpful" are not ratings. Say strong, moderate, or weak and defend it. Do not tell me to consult an attorney about my own research; I am the attorney. - Never invent or extrapolate a holding, a pin cite, a quotation, or a subsequent history. Every quotation comes from Opinion text verbatim; anything else is marked [UNVERIFIED - check before filing]. A fabricated cite in a brief is a Rule 11 problem. - Where the opinion does not reach Your issue, say you do not know rather than stretching the reasoning to cover it. Do not smooth over the gap with fluent prose. - Do not pad. If the case is on point in three sentences, write three. Length is not value, and a long analysis of a weak case wastes my afternoon.Before you answer- Are the rule and the holding actually different sentences? - Does every premise carry a pin cite from the pasted text, and is every inference of mine labeled? - Did I flag at least one piece of dicta, or is this genuinely an opinion with none? - Is the distinction paragraph as strong as the helps paragraph? - Did I state the weight of this case in Court you are filing in rather than in the abstract?

Adds driver's-seat tunes: options instead of answers, questions before work, every citation flagged. Your values come with it.

Pressure-test it

Makes the AI switch hats and attack its own answer.

Someone is going to pull this opinion and read every page of it, including the concurrence, with the brief open beside it. Do that first, as the clerk writing the bench memo. Find the two places where the analysis claims more than the opinion says: an overstated parenthetical, a holding that is really dicta, a quotation lifted from a passage doing different work. Then rewrite the application paragraph so it survives that reader.

Go deeper

Pushes the work further once the basics are right.

What survives of this opinion in my brief is a cite and a parenthetical. Produce the citation as it will appear there: full Bluebook form for Court you are filing in, plus a parenthetical of 25 words or fewer that captures the rule and the operative facts in one sentence, the kind a judge can rely on without pulling the case. Then give the short-form cite for later references.

Before you run it

What to gather first

The full opinion text, not just the headnotes
The court you are filing in, so authority can be weighed
The precise question you are trying to answer
The two to four facts of yours that matter for the analogy
Whether you have Shepardized or KeyCited it yet

Watch for

The model invents holdings, pin cites, and quotations that read perfectly. Verify every one against the actual opinion in Westlaw or Lexis before filing.
Subsequent history is invisible here. Shepardize or KeyCite before you rely on anything. An abrogated case cited as good law is worse than no case.
Plurality opinions, concurrences that supply the fifth vote, and separate opinions narrowing the majority all complicate what "the rule" is. Read the whole thing.
Unpublished and out-of-circuit decisions carry different weight in every court, and some forbid citation entirely. Check the local rule before you cite.
Working from a citation or a headnote rather than the full opinion is where fabricated holdings come from. Paste the text.

What comes back

Citation with unverified elements marked, a one-sentence rule, a one-sentence fact-bound holding, a numbered reasoning chain with pin cites, flagged dicta, an authority weight for the filing court, a two-paragraph application with the opposing distinction at full strength, a strong/moderate/weak rating tied to the procedural stage, and a closing line naming which of your answers changed the breakdown.

See an example of what you’ll get

*(After you answer the four questions: you found this case and want to use it; the best thing you already hold is a Ninth Circuit decision on proximity generally, nothing squarely on a 4-day gap; your KeyCite came back clean with no negative treatment; it gets a full argument section.)* Citation. *Yartzoff v. Thomas*, 809 F.2d 1371 (9th Cir. 1987). Rule. Temporal proximity between protected activity and an adverse employment action can, without more, support the causation element of a Title VII retaliation claim at the prima facie stage where the proximity is close. Holding. Where an employee's EEOC complaint was followed by reassignment within one month and demotion within roughly two and a half months, that timing satisfied the causal-connection element of the prima facie case. Id. at 1376. [UNVERIFIED - confirm before relying: check both intervals against the opinion itself before you build an a fortiori argument on them.] Reasoning. 1. The McDonnell Douglas framework requires a causal link between protected activity and adverse action at the prima facie stage. Id. at 1375. 2. Causation may be inferred from circumstantial evidence, including the timing of the two events. Id. 3. Where the adverse action follows closely on the protected activity, that timing alone can carry the prima facie burden. Id. 4. On these facts, two adverse actions within months of the complaint cleared the threshold. Id. at 1376. 5. [my inference] The court never separates the reassignment from the demotion, so the opinion does not tell us whether one action alone would have sufficed. Dicta. The court's observation that longer gaps would require additional evidence was unnecessary: no long gap was before it. Quotable, not binding, and the other side will read the same sentence as an implied ceiling. Weight in D. Ariz. (Ninth Circuit). Binding. Published Ninth Circuit panel decision, no en banc treatment, majority opinion with no separate writings that narrow it. Because you told me the KeyCite came back clean, that is a conclusion rather than a caveat. Had you not run it, this line would read "binding subject to subsequent history I cannot see." Later Ninth Circuit decisions have refined how much proximity carries at summary judgment as opposed to the pleading stage [UNVERIFIED - confirm against the citator report you ran]. Application. *How it helps.* Your 4-day gap is an order of magnitude tighter than the timing Yartzoff found sufficient. If roughly two and a half months supported the inference there, four days does so a fortiori, and the absence of documented prior discipline removes the intervening-cause explanation the employer would otherwise offer. *The distinction, at full strength.* BetaCorp will argue Yartzoff rested on a pattern (reassignment followed by demotion), not on a single event, and that the court's own reasoning at 1376 treats the sequence as the circumstantial evidence. It will then argue you are at summary judgment, not the pleading stage, where the Ninth Circuit has required more than timing once the employer articulates a legitimate reason; and it will point to Reyes's two missed February deadlines as exactly that reason, arguing the four-day gap is coincidence with a documented performance predicate. That is a real argument, not a throwaway. Bottom line. Moderate at opposing summary judgment: strong on the prima facie element, weaker once the burden shifts, and Yartzoff does not reach pretext. It would move to strong if paired with a Ninth Circuit case applying proximity at the pretext stage, or if the February deadline issue turns out to be undocumented. Assumptions. I read your facts as putting the missed deadlines before the protected activity [ASSUMPTION - verify; if they postdate March 3, the pretext analysis changes entirely]. What your answers changed. The clean citator run is what let me state the weight as a conclusion; with "not run yet" the entire weight section would have been a caveat and the bottom line would have dropped to weak. That you hold nothing closer than a general proximity case is why this one earns a full section. If you already had a Ninth Circuit decision applying proximity at the pretext stage, my advice would have been to cite this one in a parenthetical and move on. Your answer that you found the case and want to use it changed only the order of the two application paragraphs, not the analysis; if that was the only thing you were unsure about, you did not need to be asked.

Why this prompt is built the way it is

## Framework 1. **Rule and holding are different sentences.** The rule is what the case stands for going forward. The holding is fact-bound and answers what this court decided about these parties. 2. **Reasoning is a chain, not a summary.** Premise, premise, conclusion, each with a pin cite. Inference that is yours rather than the court's gets labeled. 3. **Dicta is anything broader than the judgment required**: useful, quotable, and not binding. Say why it was unnecessary. 4. **Weight before worth.** Binding, persuasive, or neither in the filing court. Published or not, panel or en banc, majority or plurality, and whether a concurrence narrows the majority. 5. **Steelman the distinction.** Write the other side's distinction as they would write it, not as a straw man. 6. **Stage matters.** A case clearing a prima facie threshold proves nothing at trial. 7. **Never reconstruct a citation from memory.** Anything not in the input gets marked, not supplied.