# Legal Reasoning Prover MCP > Legal Reasoning Prover forces AI agents to generate legal analysis grounded in real US law, not hallucinations. This MCP enforces strict IRAC structure, verifiable Bluebook citations, and proper jurisdiction identification. It catches fabricated case names, ensures the conclusion logically follows from the facts, and requires the agent to address the strongest counter-argument before you submit a memo. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** legal-reasoning, us-law, irac, bluebook, structured-reasoning, decision-pivots, audit, compliance ## Description Legal arguments are tough enough without your AI client inventing evidence. This connector forces structured legal reasoning across US frameworks like Bluebook citation rules and ABA Model Rules. When an agent produces a draft, it runs that text through this MCP, which audits the entire structure. It verifies if every conclusion traces back step-by-step from the rule to the facts, checking for broken syllogisms. The system also demands specific jurisdiction details—the court, the statute, or the choice-of-law basis. If the analysis is one-sided, it flags it because counsel must disclose adverse authority. Instead of just giving you a draft memo, this MCP gives you an audit report detailing exactly where the legal logic fails and what procedural remedy is needed to fix it. You connect Legal Reasoning Prover through Vinkius's catalog, making sure your AI client always adheres to professional standards. ## Tools ### validate_legal_reasoning Runs a structured audit on US legal analysis, checking for broken logic, fake citations, jurisdictional gaps, missing counter-arguments, and non-specific remedies. ## Prompt Examples **Prompt:** ``` A client was fired after reporting safety violations. Analyze whether they have a wrongful termination claim under federal law. ``` **Response:** ``` Verdict: JURISDICTION_BLIND. You state 'federal law' but don't identify which statute — SOX § 806 (public company whistleblower)? OSHA § 11(c) (workplace safety retaliation)? False Claims Act 31 U.S.C. § 3730(h) (government contractor fraud)? Each has different elements, filing deadlines, and administrative exhaustion requirements. Name the statute, the court, and whether the employee exhausted administrative remedies before filing suit. ``` **Prompt:** ``` Issue: Whether employer violated Title VII (PDA) by terminating employee after pregnancy announcement. Rule: 42 U.S.C. § 2000e; McDonnell Douglas, 411 U.S. 792 (1973). Application: Pregnancy announced March 1, PIP March 15 (no prior warnings), terminated April 1. Three comparable non-pregnant employees retained. Jurisdiction: NDIL, 7th Circuit. Counter: Employer claims performance; Ortiz v. Werner, 834 F.3d 760 (7th Cir. 2016). Remedy: EEOC charge within 300 days, back pay, damages capped $300K. ``` **Response:** ``` Verdict: REASONING_PROVEN. Syllogism sound — PDA elements mapped to facts (temporal proximity, no prior warnings, comparator retention). Authority grounded — McDonnell Douglas and Ortiz v. Werner properly cited. Jurisdiction identified — NDIL, 7th Circuit, 28 U.S.C. § 1331. Counter addressed — pretext analysis distinguishes performance defense. Remedy specific — EEOC timeline, damages cap, statutory fees. ``` **Prompt:** ``` The contract has a non-compete clause. Analyze whether it's enforceable. Courts have generally held that non-competes must be reasonable. The law supports the client's position that this is too broad. ``` **Response:** ``` Multiple rejections. (1) AUTHORITY_UNGROUNDED — 'courts have generally held' is weasel authority. Which court? Which case? (2) JURISDICTION_BLIND — non-compete enforceability varies by state. California bans them (Bus. & Prof. Code § 16600), Texas requires ancillary agreement, Illinois requires $75K salary. Which state governs? (3) SYLLOGISM_BROKEN — 'too broad' lacks element-by-element analysis of scope, duration, and restriction. ``` ## Capabilities ### Audit for Logical Gaps It checks if the conclusion genuinely follows from the law and facts provided, catching broken logical steps. ### Verify Citation Authority It confirms that every case citation is real, providing full Bluebook details to prevent hallucination. ### Pinpoint Jurisdiction It forces the agent to specify which court and law governs the dispute, eliminating vague claims like 'under US law'. ### Address Counterarguments It mandates that the agent state the strongest opposing position before proceeding with the analysis. ### Determine Actionable Remedies It requires a specific procedural action, like filing a motion or charging an agency, rather than suggesting 'consulting counsel'. ## Use Cases ### Drafting Cross-State Litigation Strategy You write a memo on contract enforceability across state lines. Your agent initially suggests 'under applicable law.' Running the text through `validate_legal_reasoning` immediately flags this as Jurisdiction Blind, forcing you to specify if it's governed by choice-of-law or Erie doctrine. ### Reviewing Whistleblower Claims Your agent outlines a whistleblower claim. The tool checks the statutes and finds that while the general theory is correct, the analysis lacks specific statutory sections (SOX § 806 or OSHA § 11(c)), requiring you to pinpoint the exact law needed for filing. ### Analyzing Employee Termination You submit a termination review based on Title VII. The MCP runs `validate_legal_reasoning` and confirms that while the core issue is covered, it failed to address potential counter-arguments regarding prior performance reviews or other adverse authority. ### Building Memos from Facts You provide raw facts for a new complaint. The MCP guides you through structuring the argument by requiring you to establish Issue, find the controlling Rule with Bluebook citations, and propose a specific procedural vehicle (like an FRCP 8 Complaint). ## Benefits - Prevents hallucinated case citations. The tool flags fabricated names and reporter volumes, ensuring only verifiably true authority makes it into your memo. - Enforces the IRAC structure rigorously. It forces the agent to map every conclusion back through Issue, Rule, Application, and Conclusion with explicit steps. - Guarantees jurisdictional specificity. You eliminate vague statements by forcing identification of the governing court, circuit, or state law basis. - Demands adversarial rigor. The MCP makes your agent confront the strongest opposing argument, providing a balanced analysis instead of one-sided theory. - Connects theory to action. Analysis must conclude with specific procedural remedies—like naming an FRCP motion type and statutory basis—not vague advice. ## How It Works The bottom line is you get an actionable audit of your legal text, showing exactly how to fix it before anyone reads it. 1. You feed the MCP your draft legal analysis—the facts and initial arguments that need vetting. 2. The MCP executes `validate_legal_reasoning`, which breaks down the text against US procedural standards (IRAC, Bluebook, etc.). 3. Your agent receives a structured report detailing every deficiency found: where the logic breaks, what citations are fake, or which jurisdiction is missing. ## Frequently Asked Questions **Does it verify if cited cases actually exist?** It catches the most common hallucination patterns — placeholder names like 'Smith v. Jones' and 'Doe v. Roe', weasel authority language like 'courts have generally held', and citations that lack proper Bluebook format. It cannot access legal databases to verify every case, but it forces the agent to provide full citation details (party names, reporter, court, year, specific holding) — which dramatically reduces fabrication because the agent must commit to verifiable specifics. **Is it only for litigation or does it work for transactional law?** The IRAC framework is most natural for litigation analysis, but the five pivots apply broadly. For transactional work: the 'syllogism' becomes 'does the contract structure achieve the stated objective?', 'authority' becomes 'is the statutory basis for this structure valid?', 'jurisdiction' becomes 'which state law governs this agreement?', and 'remedy' becomes 'what happens if the counterparty breaches?' The reasoning validation is universal — the domain language adapts. **Does it generate legal advice?** No. Legal Reasoning Prover generates zero content. It validates the STRUCTURE of legal reasoning — whether the argument follows IRAC, whether citations are properly formed, whether jurisdiction is identified, whether counter-arguments are addressed. The agent does the legal analysis; the tool proves whether that analysis is structurally sound. It is a reasoning quality gate, not a legal advisor. **What input structure does `validate_legal_reasoning` require for optimal results?** The tool requires structured legal analysis following the IRAC model (Issue, Rule, Application, Conclusion). Simply pasting raw text won't work; you must delineate specific sections to allow the agent to map facts element-by-element. **Can `validate_legal_reasoning` compare multiple state jurisdictions at once?** Yes, it forces consideration of choice-of-law principles. To analyze conflicts, you must specify the governing substantive law and define whether the analysis requires an *Erie* doctrine assessment. **If I provide ambiguous facts to `validate_legal_reasoning`, how does it handle the error?** It flags the structural deficiency. If the input premises are vague, the tool will refuse to proceed with a conclusion until you clarify the missing element or narrow the scope of the disputed facts. **Are there usage rate limits when running `validate_legal_reasoning`?** Vinkius manages API access and general rate limits apply. For high-volume legal audits, we recommend reviewing your account dashboard or batch processing through your preferred AI client. **Is the analysis from `validate_legal_reasoning` restricted solely to U.S. Common Law?** The tool is engineered specifically for US legal frameworks, enforcing Bluebook and ABA Model Rules compliance. While it can process non-US law, its required checks are limited to US jurisdiction concepts.