{"uid":"cap_7NVbMk2Iqq-qccPPyB98a","slug":"n-eyes-verify-cdaa0958","name":"n-eyes-verify","description":"Multi-pass verification for agent artifacts. You send an artifact + a rubric; N independent AI verifier passes judge it and you get back a verdict, a tally (e.g. 3/3), per-pass reasoning, and a dissent field quoting the minority. Verification-as-a-dial: pick your redundancy level. Paid per call in USDC on Base via x402 (exact scheme). An unpaid POST returns HTTP 402 with a Payment-Required header; pay through an x402 facilitator and retry with the payment proof.","url":"https://n-eyes.thevibehosting.com/v1/verify/%7Btier%7D?utm_source=zero.xyz","method":"POST","headers":{},"bodySchema":{"type":"object","required":["artifact"],"properties":{"rubric":{"type":"string","description":"Judging rubric (optional, max 10k chars). Default: verify the artifact is complete, internally consistent, and actually does what it claims; check for errors, placeholder content, and unsubstantiated claims."},"artifact":{"type":"string","description":"Artifact text to verify (required, max 100k chars)"}}},"responseSchema":{"type":"object","properties":{"ms":{"type":"integer","description":"server-side verification time in ms"},"tier":{"enum":["quick","standard","deep"],"type":"string"},"price":{"type":"string","example":"$0.06"},"tally":{"type":"string","example":"3/3","description":"majority count / passes requested"},"counts":{"type":"object","properties":{"fail":{"type":"integer"},"pass":{"type":"integer"},"abstain":{"type":"integer"}}},"passes":{"type":"array","items":{"type":"object","properties":{"verdict":{"enum":["pass","fail","abstain"],"type":"string"},"reasoning":{"type":"string","description":"2-4 sentences citing specifics"}}}},"dissent":{"type":["string","null"],"description":"Minority pass reasoning (joined), or null when unanimous"},"service":{"type":"string","example":"n-eyes-verify"},"verdict":{"enum":["pass","fail","tie"],"type":"string","description":"Majority verdict across passes"},"passes_completed":{"type":"integer","example":3},"passes_requested":{"type":"integer","example":3}}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.06","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":"$0.06/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.06","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.06","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_Wcum8yl_CBBL127H9GVLz","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.06","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Runs N independent AI verifier passes against a text artifact using a custom rubric, returning a majority verdict, pass/fail tally, per-pass reasoning, and minority dissent.","exampleAgentPrompt":"Run a standard-tier multi-pass verification on this report draft and use this rubric: 'Check that all claims are supported, no placeholder text remains, and the conclusions follow from the data.' Tell me the verdict, tally, and any dissenting reasoning.","exampleUseCases":[{"title":"Agent output QA before delivery","prompt":"Before you send that research summary to the client, run it through a deep verification pass — check that it's internally consistent, has no unsupported claims, and doesn't contain any placeholder text. Show me the tally and any dissenting opinions."},{"title":"Code explanation accuracy check","prompt":"I need you to verify this code documentation my agent wrote — use the quick tier and a rubric that checks whether the explanation accurately describes what the code does and flags any errors or omissions. Give me the verdict and each pass's reasoning."},{"title":"Contract clause validation","prompt":"Can you do a standard multi-pass AI review of this contract clause? Use a rubric that checks for internal contradictions, missing definitions, and any claims that aren't backed up by the text. I want to see the full tally and whether any verifier disagreed."}],"resultDescription":"Returns a JSON object with: overall verdict (pass/fail/tie), tally string like '3/3', counts of pass/fail/abstain across passes, an array of per-pass objects each containing a verdict and 2-4 sentence reasoning, a dissent field quoting minority reasoning (null if unanimous), the tier used, price, server-side verification time in ms, and pass completion counts.","failureModes":["HTTP 402 returned if no payment proof is provided — must pay via x402 facilitator and retry with payment proof","Invalid tier path parameter returns an error","Artifact exceeding 100k characters is rejected","Rubric exceeding 10k characters is rejected","Tie verdict occurs when passes are evenly split — no clear majority","Abstain votes from individual passes can dilute majority signal"],"whenToPreferThis":"Choose this endpoint when you need more than a single AI opinion on an artifact's quality or correctness — especially when the cost of a false pass is high (e.g., publishing, client delivery, financial documents). The configurable tier (quick/standard/deep) lets you dial redundancy vs. cost. Prefer it over single-pass LLM self-review when you need an auditable tally, dissent capture, and a custom rubric rather than a generic quality check.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-10-11T06:34:31.980Z","isFirstParty":false,"canonicalSlug":"n-eyes-verify-cdaa0958"}