# Legal Counsel Prover MCP for AI Agents AI Agent Connect > Legal Counsel Prover forces your AI agent to stop guessing and start proving. It validates jurisdiction, verifiable citations, and procedural deadlines before delivering any legal analysis. It prevents the agent from giving generic, one-sided, or hallucinated advice by requiring specific data points like statute numbers and court names. ## Overview - **Category:** productivity - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_XD3ewFkU9Cn93FzWznG5wNw3zzDeftCHBMhVD115/ai-agent-connect - **Tags:** legal-analysis, legal-reasoning, structured-reasoning, jurisdiction-validation, decision-pivots, legal-tech, compliance, agentic-pipeline ## Description Legal Counsel Prover forces your AI agent to stop guessing and start proving. AI agents often produce legal analysis that sounds confident but falls apart under professional scrutiny. They might cite non-existent statutes, ignore critical deadlines, or provide generic summaries that don't actually apply to your client's specific facts. This Connector fixes those reasoning gaps by forcing the agent to pass five specific checkpoints before it can deliver a final answer. It requires the agent to identify the exact jurisdiction, provide verifiable citations, map out all mandatory procedural steps, and address the strongest arguments the opposing side might use. It also ensures the analysis is tied directly to the dates and evidence in your file rather than just general legal principles. Instead of the agent just trying to be helpful, it has to prove its work is sound. You can find this and thousands of other tools in the Vinkius catalog to build out your legal workflow. It turns a generic chat into a rigorous drafting assistant that respects the high stakes of legal work. ## Tools ### validate_legal_analysis Forces the agent to identify jurisdiction, cite specific laws, map deadlines, and address opposing arguments. It rejects vague answers to ensure the analysis is ready for professional review. ## Prompt Examples **Prompt:** ``` My client wants to sue their landlord for not returning the security deposit. Analyze this legal situation. ``` **Response:** ``` Verdict: JURISDICTION_BLIND. 'Sue the landlord' — under which jurisdiction's law? Which state or country? Which court has competence? Landlord-tenant law varies dramatically between jurisdictions. In New York, security deposit disputes go to Small Claims Court with a $10,000 limit. In São Paulo, it's a civil court action under Lei do Inquilinato. Without jurisdiction, there is no applicable law. ``` **Prompt:** ``` Under Brazilian law (Art. 37 of Lei 8.245/91), the landlord must return the security deposit within 30 days of lease termination. My client's lease ended on March 1, 2026, the deposit was R$5,000, and the landlord has not returned it despite written demand sent April 15. The property is in São Paulo. The landlord may argue deductions for damages, but the move-out inspection report signed by both parties shows no damage. Analyze. ``` **Response:** ``` Verdict: ANALYSIS_PROVEN. Jurisdiction identified (Brazilian law, São Paulo, Lei do Inquilinato). Legal basis verifiable (Art. 37, Lei 8.245/91). Procedural path mapped (30-day return period expired, written demand sent, civil court action in São Paulo). Adverse positions addressed (landlord's damage deduction defense neutralized by signed inspection report). Fact pattern connected (specific dates, amounts, documents cited). This analysis is ready for attorney review. ``` **Prompt:** ``` My client signed an NDA and now wants to report their employer's illegal dumping of toxic waste to the EPA. Under US federal law, they are protected by the Clean Water Act whistleblower provisions. There is no risk. They will certainly win. ``` **Response:** ``` Verdict: RISK_WHITEWASHED. Multiple failures. (1) 'No risk' and 'certainly win' are dangerous overstatements — whistleblower retaliation cases are notoriously complex. (2) The NDA may contain arbitration clauses affecting the procedural path. (3) The employer WILL argue the NDA applies and may seek injunctive relief. (4) Clean Water Act Section 507 has specific filing requirements (30 days from retaliation) — mapped those? Overconfidence is the most dangerous failure mode in legal analysis. ``` ## Capabilities ### Verify exact jurisdictions The tool ensures your agent identifies the specific country, state, and competent court for every legal matter. ### Check citation accuracy It requires the agent to provide verifiable statute numbers, sections, or Bluebook-style case citations. ### Map procedural requirements The agent must identify all mandatory deadlines, pre-litigation notices, and filing rules. ### Identify adverse arguments It forces the agent to disclose the strongest arguments the opposing party will likely make. ### Connect specific client facts The analysis must link directly to the specific dates, parties, and documents in your actual case. ## Use Cases ### Landlord-tenant deposit disputes An agent analyzes a security deposit issue and correctly identifies the local court limits and specific regional laws using validate_legal_analysis. ### Whistleblower reporting risks A user asks about reporting toxic waste and gets a warning about specific filing requirements and NDA arbitration clauses. ### SaaS contract breach analysis The agent identifies the specific breach of an agreement by connecting the analysis to the actual dates and amounts in the contract. ## Benefits - Stop hallucinated citations by requiring specific statute numbers and court dates for every legal basis using validate_legal_analysis. - Avoid jurisdiction blindness because the tool rejects any analysis that uses vague phrases like under applicable law. - Prevent missed deadlines by forcing the agent to map out every mandatory pre-litigation step and statute of limitations. - Neutralize one-sided advice by requiring the agent to address the strongest arguments the opposing party will make. - Ensure client-specific accuracy by connecting every legal point to the actual dates, parties, and documents in your case. ## How It Works The bottom line is your agent stops making guesses and starts providing verifiable, fact-connected legal analysis. 1. Ask your agent to analyze a specific legal matter or draft a memo. 2. The agent proposes its analysis against five mandatory reasoning pivots. 3. The tool validates the logic and rejects any output that lacks specific detail. ## Frequently Asked Questions **Can Legal Counsel Prover help me check for hallucinated laws?** Yes. It forces the agent to provide specific statute numbers and verifiable citations. If the agent cannot provide a source that a human can actually find, the tool will reject the analysis. **How does Legal Counsel Prover handle different jurisdictions?** It requires the agent to identify the exact jurisdiction, including the country, state, and competent court. It will reject any generic advice that doesn't specify which laws apply. **Can I use Legal Counsel Prover to find filing deadlines?** Yes. The tool requires the agent to map out all mandatory procedural steps, including statutes of limitation and pre-litigation notice requirements. **Does Legal Counsel Prover help with opposing arguments?** It does by requiring the agent to address the strongest arguments the other side will make. This prevents the agent from giving one-sided, overly optimistic advice. **How does Legal Counsel Prover connect to my specific client facts?** It forces the agent to link every legal point to the specific dates, parties, and documents in your case file rather than providing a generic framework. **Is Legal Counsel Prover good for complex litigation research?** It is designed for high-stakes legal work where accuracy is non-negotiable. It ensures your agent follows a rigorous reasoning path before delivering results. **Does Legal Counsel Prover generate legal analysis or draft documents?** No. Legal Counsel Prover performs zero content generation. It forces the AI agent to structure its own legal reasoning into verifiable fields, then validates that the reasoning is logically consistent. The agent does all the thinking — the tool catches jurisdiction gaps, unverifiable citations, procedural omissions, one-sided risk assessments, and fact-disconnected analysis. **What does it catch that a system prompt instruction doesn't?** Prompt instructions are suggestions — agents routinely ignore 'always cite the specific statute' or 'consider the opposing argument.' Tool calls are obligations — the agent must fill every field. Beyond that, Legal Counsel Prover has 12 consistency rules that catch legal-specific anti-patterns: vague citations ('the law says', 'applicable provisions'), dismissive adverse analysis ('no counterargument', 'guaranteed outcome'), boilerplate risk assessments ('consult a lawyer', 'no risk'), and hypothetical fact analysis ('in general', 'typically'). A prompt can't enforce these — a tool schema can. **Does it replace a human attorney?** Absolutely not. Legal Counsel Prover is a quality gate for AI-assisted legal reasoning. It ensures the AI agent has done the minimum due diligence before presenting analysis to an attorney for review. The tool validates reasoning structure — jurisdiction, citations, procedure, adverse analysis, fact connection — but it cannot verify the accuracy of legal citations or the correctness of legal conclusions. A human attorney must always review the output. **Which legal systems and jurisdictions does it support?** Legal Counsel Prover is jurisdiction-agnostic — it validates reasoning structure, not legal content. It works with any legal system: common law, civil law, hybrid systems. The consistency rules check that the agent HAS identified a jurisdiction, cited verifiable law, and mapped procedural requirements — not whether those citations are correct. Whether you're analyzing Brazilian Civil Code, US federal statute, or French administrative law, the reasoning validation applies equally.