Upload your motion. A judicial panel shows you exactly how it fails — before opposing counsel does.
PreRuling shows you how five judges will rule, where your motion breaks, and the smallest change that improves your odds.
If you're filing one of these, PreRuling was built for you.
Upload your filing. A calibrated panel rules on it, finds the argument most likely to kill it, and shows you the smallest change that flips the outcome — in about two minutes.
The one argument that sinks your motion — surfaced before opposing counsel finds it.
Exactly how the other side will attack — so you answer it in your brief, not at the hearing.
Every citation checked against a live 18M-record mirror — fabricated cases caught before you file.
Watch your odds move as you apply each fix — before a word reaches the court.
The smallest set of edits that changes the ruling — not a rewrite.
A confidence grade on the whole read — so you know when it's safe to file.
PreRuling wasn't tuned on hypotheticals. It was built and pressure-tested inside active federal litigation — run on real filings before they reached the clerk.
Every citation is checked against a live 18M+ citation mirror — existence confirmed, or flagged, before you file.
Every motion has one argument that does the most damage to it. We find it first — and show you how to answer it before opposing counsel ever makes it.
Why pay $1,200+ for one associate's read when you can run five judges — with citation verification and opposition strategy — for less than a cup of coffee?
Run Your Motion →Run your motion through the panel and you walk away with a sharper filing: the weak element identified, the argument opposing counsel would have made surfaced and countered, every citation checked, and the smallest change that improves your odds spelled out.
It works the same whether you're refining a draft or pressure-testing something you're about to file. Better odds before you spend the filing fee.
Already filed? Run it anyway — you'll see exactly what opposing counsel is about to argue, in time to prepare your reply.
Party names are locked. Theories are selected. Jurisdiction is set. The panel cannot drift, flip parties, or apply the wrong standard.
Name your plaintiff and defendant. Every judge prompt is hardcoded: these roles are fixed and cannot be reassigned. Allegations are never treated as established facts.
Select all theories that apply. Each evaluated independently. Panel flags theory drift if a judge strays outside your selected theories.
Select your district. Optionally select your assigned judge — D. Nev. profiles cover five documented judicial philosophies across the district. The entire panel calibrates to their ruling patterns.
Tag what you're submitting. Initial Complaint triggers Twombly/Iqbal. MTD triggers 12(b)(6) attack. Opposition Brief checks whether it successfully rebuts.
Your theory selection determines which judges hear your motion. Civil rights gets civil rights specialists. Consumer law gets consumer law specialists.
Three levels of scrutiny. The more judges, the more pressure — and the more intelligence you walk away with.
Each judge has a distinct philosophy. Your theory selection determines who hears your motion.
Every simulation delivers a complete litigation intelligence package. Here's exactly what's in it — sourced from the live codebase.
Plaintiff and defendant names hardcoded into every judge prompt. Roles cannot flip. Allegations are never treated as established facts.
Each carries its own legal standard: Preliminary Injunction (Winter four-factor), MTD (Twombly/Iqbal), Summary Judgment, Sanctions, Default Judgment, TRO, Motion to Compel, Attorney Fees, Stay, Reconsideration, Strike, and Other.
Federal District, Federal Appellate, State General, State Appellate, State Limited, and Bankruptcy Court. Circuit precedent overlays applied automatically. Bankruptcy triggers automatic stay analysis.
FHA § 3617, FHA § 3604, FDCPA, FCRA, ECOA, § 1983, Breach of Contract, Fraud, Negligence, Abuse of Process, Slander of Title, Tortious Interference, DTPA, Rule 11, and Other. Theory routes your motion to a calibrated judge panel.
Tag what you're submitting: Initial Complaint, MTD, Opposition Brief, Reply Brief, Cause of Action Only, or Full Document. Each triggers a different analytical standard in every judge prompt.
Paste directly for clean, complete ingestion. Bypasses PDF OCR entirely. Live word count and capacity percentage per slot. Chunked mode auto-triggers for large complaints at 95K characters.
Issues that independently kill the motion. Hard Fatal (do not file), Soft Fatal (fix first), Tactical (strengthen). Each includes a Minimum Change to Win.
The single most lethal argument against this motion — surfaced from the most skeptical judge and displayed first. If you can't answer it, you're not ready to file.
Each judge identifies the argument most likely to defeat this motion. All five displayed together — shows where the panel converges on your weakest point.
When the panel splits, PreRuling extracts the exact reasoning on both sides. Why it grants. Why it denies. Use the grant logic as your primary argument. The deny logic is opposing counsel's attack — you see it coming.
Each judge evaluates your Primary Theory, an Alternative Theory if primary fails, and an independent Equitable Basis. The motion succeeds if any single theory independently meets the standard.
If a judge analyzes outside your selected legal theory, a warning banner fires on their ruling card. You selected the theory — the panel stays in it.
What opposing counsel will argue per judge philosophy — DEVASTATING, STRONG, or MODERATE lethality. Counter-reframe per argument. Killer Bench Question per judge.
Inside Opposition Intel™ — identifies specific defense framings in your own filing, exposure level, and the exact counter each requires. Shows how they'll use your own language against you.
Two analyses per judge, never blended. Strict shows the likely outcome. Strategic shows the Key Reframe. Toggle between Survive (can this hold?) and Win (what does it take to win outright?).
Inside Win mode — each judge outputs a specific strategy, required changes, and an estimated outcome shift percentage. A specific path to GRANTED with probability attached.
Every weakness ranked by severity and confidence. Each patch includes the smallest single edit that neutralizes the issue. LOW confidence patches flagged with a warning.
Alternative legal theories that survive even when strict analysis says deny. Viability ratings (HIGH / MEDIUM / LOW) and exactly which weakness each path overcomes.
For each judge: Primary Obstacle, Flip Theory, numbered Minimum Changes Required, whether new evidence is needed, and confidence if flipped.
Run Outcome Shift™ — Apply All Patches. Reruns the full panel with patches implemented. Shows which judges flipped and how confidence changed.
Live in the sticky header. Factors in judge success rate, model collapse detection, volatility, and average confidence. A means file it. F means fix your inputs first.
Every citation checked against 18M+ citations from a 70M+ record legal corpus. Real authorities confirmed, fabricated ones flagged — the safeguard against filing case law that doesn't exist.
Detects if judges produce identical reasoning or confidence scores. Flags it in the header and penalizes the Panel Reliability Score automatically.
Download a complete HTML report — all rulings, fatal flaws, patches, opposition intel, and survival paths. Attorney-ready and shareable.
Upload your motion. See where it's weak, see the argument the other side will make, see every citation confirmed, and see the smallest change that improves your odds. Then file it sharper than you would have.
PreRuling checks each citation in the panel output against a CourtListener mirror — 18M+ citations across a 70M+ record legal corpus, 3,300+ jurisdictions. Real authorities get confirmed; anything fabricated gets flagged before it reaches your filing. This is what separates PreRuling from a generic AI tool that simply invents case law.
Confirmed present in CourtListener — PreRuling's citation verifier checks existence against a CourtListener bulk-data mirror. Covers federal statutes, the Federal Rules, reporters, and major NRS sections. Existence only — confirm current validity before filing.
Not located in the mirror. May be more recent than the export, unpublished, or miscited. Not a verdict on validity — verify directly before relying on it.
Looks like a citation but matches known hallucination patterns and isn't in the mirror. Treat as potentially fabricated. Do not include in any court filing without independent verification.
An actual Panel of 3 run on a real First Amended Complaint (ADA Title III, NRS 651, battery, fraud, civil conspiracy). The motion to dismiss was DENIED — the complaint survives — and the panel still surfaced the one issue worth fixing first.
What takes weeks of briefing and oral argument happens here in minutes.
Define the case before the panel sees a single word. Name the parties, select your legal theories, choose your court, and optionally select your assigned judge. This context is locked into every judge prompt — no drift, no party flipping, no wrong standard.
Upload PDFs to the three-column layout (Plaintiff / Motion / Defendant) or paste directly for 100% ingestion. Large complaints are automatically chunked at logical section breaks. A live size meter shows exactly how much is being analyzed.
An interpretation layer runs first — analyzing timeline, irreparable harm, status quo, and contradictions. Each theory-calibrated judge rules independently. Results stream in progressively. Panel Reliability Score A–F and model collapse detection run automatically.
Review Fatal Flaws, apply patches, run Outcome Shift™, check Flip This Ruling for DENIED judges. Export the full report. File with confidence — or don't file at all. That's the point.
Jurisdiction-specific case law overlays for 30+ federal districts and major state courts. Select your district and the panel applies dominant precedents and procedural quirks for that court.
An associate reviewing one motion for weaknesses runs $600–$2,000 and still can't simulate a five-judge panel or hand you opposition intel. PreRuling does more, in minutes. One denied motion, one sanctions exposure, one kill shot you didn't see costs more than a year of this. We host the engine — nothing to set up, no API key to manage.
PreRuling was not built in a demo environment. It was used to armor real filings in active federal and state cases before they were submitted — pressure-testing live motions against the panel, surfacing the kill shots, and closing the gaps opposing counsel would have exploited. The tool earns its price on a single motion you would otherwise have filed blind.
We're finalizing credit pricing and the monthly plan tiers. Checkout reopens in about a week. Check back soon to run your first motion through the panel.