{"uid":"cap_VI9kdpKvC17NewbQlRSoZ","slug":"verity-suite-arbiter-pro-582504e4","name":"Verity Suite Arbiter Pro","description":"The trust fabric for AI agents — calibrated, fail-closed services agents pay per call.","url":"https://suite.veritylayer.dev/arbiter/pro","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"claims":{"type":"array","items":{"type":"string"},"title":"Claims","description":"Two or more mutually conflicting claims/positions to adjudicate, one per item, in stable order. Fewer than two means nothing to adjudicate."},"evidence":{"type":"array","items":{"type":"string"},"title":"Evidence","description":"Retrieved evidence excerpts/sources to judge from. If empty, weak, or off-topic, the verdict MUST reflect that (insufficient_evidence)."},"question":{"anyOf":[{"type":"string","maxLength":2000},{"type":"null"}],"title":"Question","default":null,"description":"The specific point in dispute the claims are answering, if narrower than the claims themselves. Used to scope relevance only."}}},"responseSchema":null,"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.35","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.35/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.35","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.35","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_NmECbGTqypOFKcYdyZcPF","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.35","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Adjudicates mutually conflicting claims against provided evidence, returning a signed verdict with confidence score and cited reasons","exampleAgentPrompt":"I have two conflicting claims about climate policy — 'Renewable energy now supplies over 30% of global electricity' vs 'Renewables account for less than 20% of global electricity' — and I've pulled three evidence excerpts from recent IEA reports. Can you adjudicate which claim is better supported and give me a signed verdict with your reasoning?","exampleUseCases":[{"title":"RAG pipeline contradiction resolver","prompt":"My retrieval-augmented pipeline produced two conflicting answers: one says the FDA approved drug X in 2021, another says 2023. Here are the source passages I retrieved — can you adjudicate which claim is better supported by the evidence and tell me why?"},{"title":"Fact-check duel between news sources","prompt":"Two news articles disagree: one claims GDP growth was 3.2% last quarter, the other says 2.1%. I have the original government data excerpts — can you rule on which claim the evidence actually supports and give me a signed receipt I can log?"},{"title":"Agent debate arbiter for research workflow","prompt":"I have two AI-generated research summaries making opposite claims about whether coffee increases cardiovascular risk, and I've gathered five study excerpts. Please adjudicate the dispute — I need the verdict, your confidence level, and specific reasons citing which evidence supports or undercuts each claim."}],"resultDescription":"Returns a JSON object with: a verdict enum (claim_supported, undecidable, or insufficient_evidence), a confidence score, an array of concrete reasons citing specific evidence items by content and claim index, and an optional Ed25519-signed VerityLayer receipt for tamper-evident audit logging.","failureModes":["Fewer than two claims provided — nothing to adjudicate, returns error","Evidence array is empty or off-topic — verdict will be insufficient_evidence","Claims are not mutually contradictory — adjudication may be undecidable","Evidence excerpts are truncated or paraphrased — may reduce confidence or yield undecidable","Signing not configured server-side — receipt field returns null","Malformed input schema (missing required fields) — HTTP 4xx error"],"whenToPreferThis":"Choose this endpoint when you need a structured, calibrated, fail-closed ruling on a factual dispute between exactly two or more conflicting claims, especially when you have retrieved evidence and need a citable, optionally signed verdict. Prefer it over generic LLM prompting when auditability (signed receipt), determinism, and explicit confidence scoring matter — for example in RAG pipelines, multi-agent debate resolution, compliance workflows, or any context where the adjudication must be loggable and verifiable. Not appropriate for open-ended summarization, content generation, or single-claim fact lookup.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T06:41:39.508Z","isFirstParty":false}