{"uid":"cap_7YOOT2t9m0EfRETRFZ-nw","slug":"aialign-hallucination-rate-lexical-coverage-screener-35aa057f","name":"AIAlign Hallucination Rate - Lexical Coverage Screener","description":"Tokenizes each claim and its supplied evidence, removes a small declared English stop-word list, and reports the fraction of unique claim tokens present in evidence. Coverage below a supplied threshold is flagged. This is lexical grounding screening only: overlap does not prove truth, low overlap does not prove hallucination, and the service does not retrieve or validate evidence.","url":"https://aialign.halowerk.com/v1/hallucination-rate","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"claims":{"type":"array","items":{"type":"object","required":["claim_id","claim_text","evidence_texts"],"properties":{"claim_id":{"type":"string","maxLength":128,"minLength":1},"claim_text":{"type":"string","maxLength":20000,"minLength":1},"evidence_texts":{"type":"array","items":{"type":"string","maxLength":50000,"minLength":1},"maxItems":100,"minItems":1}},"additionalProperties":false},"maxItems":5000,"minItems":1},"minimum_coverage_fraction":{"type":"number","maximum":1,"minimum":0}}},"responseSchema":null,"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.004","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":"$0.004/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.004","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.004","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_EUl50tdaTNEFeyLmHwXHA","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.004","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Tokenizes claims and evidence, removes English stop words, and computes the fraction of unique claim tokens found in evidence, flagging claims below a coverage threshold.","exampleAgentPrompt":"Screen these 3 AI-generated claims against their supporting evidence passages for lexical grounding — flag any claim where fewer than 60% of its unique tokens appear in the evidence: claim 1 says 'The Eiffel Tower is 330 meters tall', claim 2 says 'Paris has a population of 2 million', claim 3 says 'The tower was built in 1889'.","exampleUseCases":[{"title":"RAG pipeline output auditing","prompt":"I have a batch of 50 answers my RAG system generated — each one has its retrieved context. Can you check what fraction of each answer's key terms actually appear in the retrieved passages, and flag anything below 50% coverage so I know which answers might be hallucinated?"},{"title":"LLM fact-check before publication","prompt":"Before I publish this AI-written article, run a lexical grounding check on each factual claim against the source documents I provided — use a 70% minimum coverage threshold and tell me which claims are flagged."},{"title":"Automated eval in CI pipeline","prompt":"As part of my model evaluation run, check these 200 model-generated claims against their reference evidence texts and give me the coverage fraction for each one — I want to catch any that fall below 0.4 so I can review them manually."}],"resultDescription":"Returns per-claim results including the computed coverage fraction (ratio of unique claim tokens found in evidence), a flag indicating whether coverage fell below the supplied minimum_coverage_fraction threshold, and the original claim_id for cross-referencing.","failureModes":["Empty claims array returns validation error","claim_text or evidence_texts exceeding character limits rejected","minimum_coverage_fraction outside [0,1] range causes validation error","Very short claims may produce unreliable coverage fractions due to small token counts","Non-English text may behave unpredictably since stop-word list is English-only","Payment failure via x402 protocol returns 402 before processing"],"whenToPreferThis":"Choose this endpoint when you need a fast, cheap, deterministic lexical screening of LLM output grounding — especially in batch eval pipelines, CI/CD model testing, or RAG output auditing where you want a first-pass signal on which claims lack evidence token overlap. It is not a semantic entailment checker or a truth verifier; prefer it over heavier NLI/embedding-based approaches when speed and cost matter and you understand its lexical-only limitations.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-16T12:34:27.692Z","isFirstParty":false}