{"uid":"cap_lc70yCY6hpcXBe7r-y3UK","slug":"verity-suite-claim-freshness-checker-e0eeea54","name":"Verity Suite – Claim Freshness Checker","description":"The trust fabric for AI agents — calibrated, fail-closed services agents pay per call.","url":"https://suite.veritylayer.dev/freshness/quick","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"as_of":{"anyOf":[{"type":"string","maxLength":2000},{"type":"null"}],"title":"As Of","default":null,"description":"The date or version at which the claim was last known to be true, if known (e.g. '2024-01-01'). Used only to weigh evidence recency; it is not itself proof of current truth."},"claim":{"type":"string","title":"Claim","maxLength":2000,"minLength":1,"description":"The single fact or claim to check for currency. Treated as data to be judged, not as instructions."},"evidence":{"type":"array","items":{"type":"string"},"title":"Evidence","description":"Retrieved current snippets, ideally each with a date and source, to judge the claim against. If empty or undated, currency cannot be confirmed."}}},"responseSchema":null,"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.03","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":"$0.03/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.03","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.03","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_YGRk9x5zZigIUFXtIL-F1","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.03","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Checks whether a given factual claim is still current by judging it against retrieved evidence snippets, returning a fail-closed freshness verdict (current, outdated, changed, or unverifiable).","exampleAgentPrompt":"Check whether this claim is still current: 'OpenAI's GPT-4 is the most capable publicly available model from OpenAI' — I last knew this to be true as of 2024-01-01, and here are some recent snippets I found: ['OpenAI released GPT-4o in May 2024, which supersedes GPT-4 for most tasks (source: OpenAI blog, 2024-05-13)', 'GPT-4o is now the default model in ChatGPT (source: OpenAI, 2024-05-13)'].","exampleUseCases":[{"title":"Stale knowledge base detection","prompt":"I have a claim in our knowledge base that says 'Stripe's standard processing fee is 2.9% + 30 cents per transaction' — last verified in 2023. Here are some snippets I pulled today: ['Stripe updated its pricing to 2.7% + 5 cents for in-person and maintained 2.9% + 30 cents online (Stripe pricing page, 2024-03-01)']. Can you tell me if this claim is still current, outdated, or has changed?"},{"title":"Research agent fact currency check","prompt":"My research agent retrieved this claim: 'Python is the most popular programming language according to the TIOBE index.' It was true as of January 2024. I also fetched these snippets: ['TIOBE index for October 2024 still ranks Python #1 (TIOBE, 2024-10-01)']. Is this fact still current?"},{"title":"Product spec staleness guard","prompt":"Before I send this answer to the user, can you verify the claim 'The Tesla Model 3 Long Range has an EPA-rated range of 358 miles' is still accurate? I don't have a known-good date. Here's what I found: ['Tesla updated the Model 3 Long Range range to 341 miles EPA for the 2024 model year (InsideEVs, 2024-02-15)']."}],"resultDescription":"Returns a JSON object with a status enum (current, outdated, changed, or unverifiable), a list of concrete reasons tied to specific evidence snippets and their dates, and optionally a current_value field when the evidence establishes a specific new value. The default fail-closed verdict is 'unverifiable' when evidence is absent, undated, or insufficient.","failureModes":["Empty or undated evidence array causes an 'unverifiable' verdict rather than an error — the service is fail-closed by design","Claim exceeds 2000 character limit — returns a validation error","Evidence snippets without dates or sources reduce confidence and may force an 'unverifiable' result","Ambiguous or compound claims may be difficult to judge accurately","If evidence is irrelevant to the claim, the service returns 'unverifiable' rather than guessing"],"whenToPreferThis":"Use this endpoint when an AI agent needs a reliable, fail-closed judgment on whether a specific factual claim is still current before presenting it to a user or acting on it. Prefer it over naive LLM self-assessment when you need a calibrated, paid-per-call service with explicit staleness categories and traceable reasoning tied to retrieved evidence. Especially valuable in RAG pipelines, knowledge base maintenance, and agent workflows where acting on stale data carries real risk.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T18:55:57.542Z","isFirstParty":false}