Every change Wellknown observed on this MCP server, newest first, with what it was before and what it became. Tool-surface changes carry the definition diff. Nothing here is edited after the fact.
Changed the definition of "get_innovation_flags", "get_updates", "query_claims" and 2 more
⟨531 unchanged words⟩ ,"type":"integer"},"total":{"titleanyOf":[{"Totaltype":"integer"},{"type":"integernull"}],"description":"Matching claims for this query when that count is known. Null when the page was read without a count. Null is the same on every page of that query. It is not the number of rows returned.","title":"Total"}},"required":["limit","offset" ⟨142 unchanged words⟩
⟨388 unchanged words⟩ ,"type":"integer"},"total":{"titleanyOf":[{"Totaltype":"integer"},{"type":"integernull"}],"description":"Matching claims for this query when that count is known. Null when the page was read without a count. Null is the same on every page of that query. It is not the number of rows returned.","title":"Total"}},"required":["limit","offset" ⟨315 unchanged words⟩
⟨388 unchanged words⟩ ,"type":"integer"},"total":{"titleanyOf":[{"Totaltype":"integer"},{"type":"integernull"}],"description":"Matching claims for this query when that count is known. Null when the page was read without a count. Null is the same on every page of that query. It is not the number of rows returned.","title":"Total"}},"required":["limit","offset" ⟨307 unchanged words⟩
⟨117 unchanged words⟩ ,"type":"integer"},"total":{"titleanyOf":[{"Totaltype":"integer"},{"type":"integernull"}],"description":"Matching claims for this query when that count is known. Null when the page was read without a count. Null is the same on every page of that query. It is not the number of rows returned.","title":"Total"}},"required":["limit","offset" ⟨47 unchanged words⟩
⟨388 unchanged words⟩ ,"type":"integer"},"total":{"titleanyOf":[{"Totaltype":"integer"},{"type":"integernull"}],"description":"Matching claims for this query when that count is known. Null when the page was read without a count. Null is the same on every page of that query. It is not the number of rows returned.","title":"Total"}},"required":["limit","offset" ⟨570 unchanged words⟩
Certificate recorded, valid to 2027-01-14
Authorization not required
Changed the definition of "get_innovation_flags", "get_updates", "query_claims" and 1 more
⟨548 unchanged words⟩ ":{"description":"Why two claims werelinked.linked.\n\nAnOlderolderflagsstoredomitrow omits this field andloadstill loads asnull.\n\nsame_sourcenull. Scoring\nnever returns null. ``reclassify_stored_flags`` writes a basis onto\nevery row it can score, and writes lexical onto a null row it cannot\nscore. That fallback
Changed the definition of "get_evidence", "get_innovation_flags", "get_updates" and 2 more
Provenance for one claim. Excerpts are short unless full_text is true. unpublished is true for a private unpublished paper. The stored excerpt, page, and section stay on provenance. status_note on a consensus claim with no stored contradiction is: No stored contradiction; not independent multi-source consensus. evidence_note is: Extractor confidence in the claim statement, not study quality. integrity_warning is set when the statement unit looks truncated relative to the excerpt, or the magnitude looks extreme relative to the excerpt. The statement and the excerpt are not rewritten. Use this after a search when you have ⟨20 unchanged words⟩
⟨107 unchanged words⟩ "default":null,"title":"Domain"},"evidence_note":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Extractor confidence in the claim statement, not study quality. evidence_strength is that extractor confidence.","title"
Changed the definition of "get_innovation_flags"
List innovation flags, strongest relationship first. Use this when you want links between claims rather than the claims themselves. Every flag includes flag_type, cross_domain, and summary. summaryis the short reason tosaysfollowwhy thepair orclaimsskipareit.linked. flag_type is lexical forasharedkeywordterminology with no substantive relation,cross_domain
Changed the definition of "get_evidence", "get_innovation_flags", "get_updates" and 2 more
⟨63 unchanged words⟩ ],"default":null,"title":"Section"},"url":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"title":"Url"}},"required":["excerpt"],"title": ⟨98 unchanged words⟩
⟨10 unchanged words⟩ links between claims rather than the claims themselves.EachEvery flag includesflag_type.flag_typelexical_association
Changed the definition of "get_innovation_flags", "query_claims" and "search_smart"
⟨10 unchanged words⟩ links between claims rather than the claims themselves. Each flag includes flag_type. lexical_association is exploratory shared-term context. scientific_connection and research_hypothesis are evaluated relationships. Returns flag id, domain, summary, strength,andflag_type, connected_domain, the claim ids on eachpath.path, and a claims array in that same order. Each claims item has claim_id, statement, and evidence strength. Full evidence stays on get_evidence. connected_domain is the primary domain of the claim that does not match the flag domain. It is null when the claims share a primary domain. Optional cross_domain_only, when true, returns only flags whose claims sit in different primary domains. Repeated claim paths return once, keeping the strongest flag for that pair. Example: get_innovation_flags(domain="ai-ml", min_strength=0.5, limit=10). Example: get_innovation_flags(domain="quantum", min_strength=0.8, limit=5). Example: get_innovation_flags(domain="consciousness", cross_domain_only=True, limit=10
Unknown → Live
First tool surface recorded: 8 tools (server version 0.1.0)
https://claimora.si/mcp (mcp_streamable_http) — from mcp_registry, with the record
Showing the latest 11 events. The API returns up to 500 and filters by kind: ?kind=tool_surface_changed
⟨570 unchanged words⟩ ":{"description":"Why two claims werelinked.linked.\n\nAnOlderolderflagsstoredomitrow omits this field andloadstill loads asnull.\n\nsame_sourcenull. Scoring\nnever returns null. ``reclassify_stored_flags`` writes a basis onto\nevery row it can score, and writes lexical onto a null row it cannot\nscore. That fallback is explicit: the enum has no conceptual value, so\noverlap that is not a bound technique, a stated transfer, a stored\ncontradiction, or the same paper is lexical. It is not upgraded.\n\nsame_source is the same paper and is not independent evidence.\nlexical is sharedvocabulary.vocabulary,methodologicalincluding a generic label.\nmethodological is a technique boundtechnique.\ntransferto a shared specific noun.\ntransfer is a mechanism the claims state. contradiction is ⟨40 unchanged words⟩
⟨570 unchanged words⟩ ":{"description":"Why two claims werelinked.linked.\n\nAnOlderolderflagsstoredomitrow omits this field andloadstill loads asnull.\n\nsame_sourcenull. Scoring\nnever returns null. ``reclassify_stored_flags`` writes a basis onto\nevery row it can score, and writes lexical onto a null row it cannot\nscore. That fallback is explicit: the enum has no conceptual value, so\noverlap that is not a bound technique, a stated transfer, a stored\ncontradiction, or the same paper is lexical. It is not upgraded.\n\nsame_source is the same paper and is not independent evidence.\nlexical is sharedvocabulary.vocabulary,methodologicalincluding a generic label.\nmethodological is a technique boundtechnique.\ntransferto a shared specific noun.\ntransfer is a mechanism the claims state. contradiction is ⟨32 unchanged words⟩
⟨570 unchanged words⟩ ":{"description":"Why two claims werelinked.linked.\n\nAnOlderolderflagsstoredomitrow omits this field andloadstill loads asnull.\n\nsame_sourcenull. Scoring\nnever returns null. ``reclassify_stored_flags`` writes a basis onto\nevery row it can score, and writes lexical onto a null row it cannot\nscore. That fallback is explicit: the enum has no conceptual value, so\noverlap that is not a bound technique, a stated transfer, a stored\ncontradiction, or the same paper is lexical. It is not upgraded.\n\nsame_source is the same paper and is not independent evidence.\nlexical is sharedvocabulary.vocabulary,methodologicalincluding a generic label.\nmethodological is a technique boundtechnique.\ntransferto a shared specific noun.\ntransfer is a mechanism the claims state. contradiction is ⟨295 unchanged words⟩
List innovationflags,flags.strongestTherelationshipdefaultfirst.sort is discovery. Use this when you want links between claims rather than the claims themselves. Pass claim_id to navigate from one claim to the flags that contain it. Each hit includes flag_type, strength, summary, the other claim ids, and those claims' primary domains. Filters are flag_type, min_strength, and cross_domain_only. The claim list is ranked by relationship strength and novelty. A path that only names the claim, a partner that repeats the claim, and a near-duplicate of a partner already returned are omitted. Discovery combines relationship strength, novelty, and conceptual overlap. Near-duplicate claims rank lower on the first page and remain in the list. sort strength orders by the stored relationship strength. Every flag includes flag_type, cross_domain, complementary, and summary. summary says why the claims are linked. complementary is a ranking hint when two independent sources share a measurable variable or process and discuss different aspects. It reuses relationship_basis, does not change flag_type, and does not state a hypothesis. A weak link stays weak. flag_type islexicallexical_association for shared terminology with no substantive relation, ⟨7 unchanged words⟩ contradiction for a stored disagreement, or innovationforwhen the claims state anamedmechanismtransferthatorcouldextension.transfer. lexical selects the same weak rows. cross_domain is true when the claims have different primary domains. It is not aflag_type.flag_type and does not raise strength. Older rows may still uselexical_association,cross_domain,or research_hypothesis. strength isthecomputedsubstanceforofthatthepair.link.It is not claim evidence strength. A weak lexical linkis low. Itstaysinlowtheandresultremains when min_strength is omitted or 0. Returns flag ⟨72 unchanged words⟩ flag_type keeps one label, for example lexical. Optional sort is discovery or strength. discovery is the default. Repeated claim paths return once, keeping the strongest flag for that pair. Example: get_innovation_flags(domain="ai-ml", min_strength=0.5, limit=10). Example: get_innovation_flags(domain="ai-ml", sort="strength", limit=10). Example: get_innovation_flags(claim_id="claim-attention", min_strength=0.5, limit=10). Example: get_innovation_flags(domain= ⟨13 unchanged words⟩
{"properties":{"claim_id":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Return flags whose path contains this claim. Example: claim-attention.","examples":["claim-attention"],"title":"Claim Id"},"cross_domain_only":{"anyOf":[{"type":"boolean"},{"type":"null"}],"default":falsenull,"description":"When true, return only flags whose connected_domain is set.DefaultNull and the default are false.","examples":[false],"title":"Cross Domain Only","type":"boolean"},"domain":{"anyOf":[{"type" ⟨23 unchanged words⟩ default":null,"description":"Keep one flag_type.Example:lexicallexical.and lexical_association are the same weak class. Empty returns every type.","examples":["lexical"],"title":"Flag Type"},"limit":{"anyOf":[{"maximum":50,"minimum":1,"type":"integer"},{"type":"null"}],"default":20null,"description":"Page size. Example: 10. Null uses 20.","examples":[10],"maximum":50,"minimum":1,"title":"Limit","type":"integer"},"min_strength":{"anyOf":[{"maximum"
⟨33 unchanged words⟩ ":"Claims","type":"array"},"complementary":{"default":false,"description":"Ranking hint. True when two independent sources share a specific measurable variable or process and discuss different aspects. It reuses relationship_basis, does not change flag_type, and does not state a hypothesis. A weak lexical link stays weak.","title":"Complementary","type":"boolean"},"connected_domain":{"anyOf":[{"type":"string ⟨65 unchanged words⟩ lexical_association","description":"Kind of link. lexicalisand lexical_association are shared terminology with no substantive relation. scientific_connection is a conceptual or methodological relationship. innovation is anamedtransferorwhoseextension.mechanism is stated in the claims. contradiction is a stored disagreement. Older rows may still saylexical_association,cross_domain,or research_hypothesis."},"id":{"title":"Id","type":"string"},"near_duplicate":{"default":false,"description":"True when the claim statements are near-duplicates. The row stays in the list. Discovery ranking places it lower.","title":"Near Duplicate","type":"boolean"},"other_claim_ids":{"description":"Claim ids on this path other than the claim_id passed to get_innovation_flags. Empty when that argument is omitted.","items":{"type":"string"},"title":"Other Claim Ids","type":"array"},"other_domains":{"description":"Primary domain of each other claim, in other_claim_ids order. Empty when claim_id is omitted.","items":{"type":"string"},"title":"Other Domains","type":"array"},"relationship_basis":{"anyOf":[{"$ref":"#/$defs/RelationshipBasis"},{"type":"null"}],"default":null,"description":"Why the claims are linked. same_source is the same paper and is not independent evidence. lexical is shared vocabulary. methodological is a bound technique. transfer is a mechanism the claims state. contradiction is a stored disagreement. An older row is filled on read when the claim text supports one of these. A conceptual pair with no such signal stays null. Lexical overlap is not relabeled as methodological or transfer."},"
⟨27 unchanged words⟩ evidence strength, a shortened excerpt, and source_url,whendoi,aandpaperpaper_idlinkwhenisthose are known. Example: get_updates(since="2026-01-01T00:00:00Z" ⟨14 unchanged words⟩
⟨24 unchanged words⟩ "One stored contradicting claim, when one ispresent.present,claim_idwith the same compact source pointer."},"doi":{"default":"","description":"StoredstatementDOI,andthestrength.DataCite DOI of an arXiv id, or a bioRxiv paper id that is already a DOI. Empty when unknown. Full provenance stays on get_evidence.","title":"Doi","type":"string"},"domain":{"title":"Domain","type":"string"},"evidence_note":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Extractor confidence in the claim statement, not study quality. evidence_strength is that extractor confidence.","title":"Evidence Note"},"evidence_strength":{"title":"Evidence Strength","type":" ⟨8 unchanged words⟩ ":"Id","type":"string"},"integrity_warning":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Set when the statement unit looks truncated relative to the excerpt, or the magnitude looks extreme relative to the excerpt. The statement and the excerpt are not rewritten.","title":"Integrity Warning"},"match_reason":{"default":"","description":"Why this hit ranked here: keyword overlap, domain match, andevidenceretrievalstrength.relevance. evidence_strength is the stored score and is named here. The number stays on evidence_strength.","title":"Match Reason","type":"string"},"paper_id":{"default":"","description":"Stored paper id when one is known. A private paper is an opaque private-paper id. Empty when unknown. Full provenance stays on get_evidence.","title":"Paper Id","type":"string"},"source_url":{"default":"","description":" ⟨31 unchanged words⟩ ,"status":{"$ref":"#/$defs/ClaimStatus"},"status_note":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Set on a consensus claim with no stored contradiction: No stored contradiction; not independent multi-source consensus. Null for contested, outdated, or a claim that already has a contradiction.","title":"Status Note"},"supporting":{"description":"Up to two other claims from the same papers. Each has claim_id, statement, strength, andstrength.the same compact source pointer.","items":{"$ref":"#/$defs/RelatedClaim"},"title":"Supporting","type":"array"},"unpublished":{"default":false,"description":"True when this claim is from a private unpublished paper. Public sources stay false.","title":"Unpublished","type":"boolean"}},"required":["id","statement","domain ⟨60 unchanged words⟩ "Claim Id","type":"string"},"doi":{"default":"","description":"Stored DOI, the DataCite DOI of an arXiv id, or a bioRxiv paper id that is already a DOI. Empty when unknown.","title":"Doi","type":"string"},"paper_id":{"default":"","description":"Stored paper id. A private paper is an opaque private-paper id. Empty when unknown.","title":"Paper Id","type":"string"},"relationship_basis":{"anyOf":[{"$ref":"#/$defs/RelationshipBasis"},{"type":"null"}],"default":null,"description":"Why this related claim is shown. same_source means the same paper and is not independent evidence. lexical, methodological, transfer, and contradiction name the other stored signals. Null when unknown."},"source_url":{"default":"","description":" ⟨27 unchanged words⟩ "title":"Strength","type":"number"},"unpublished":{"default":false,"description":"True when that claim is from a private unpublished paper.","title":"Unpublished","type":"boolean"}},"required":["claim_id","statement","strength"],"title":"RelatedClaim","type":"object"},"RelationshipBasis":{"description":"Why two claims were linked. Older flags omit this and load as null.\n\nsame_source is the same paper and is not independent evidence.\nlexical is shared vocabulary. methodological is a bound technique.\ntransfer is a mechanism the claims state. contradiction is a stored disagreement.","enum":["same_source","lexical","methodological","transfer","contradiction"],"title":"RelationshipBasis","type":"string"}},"properties":{"claims":{"items ⟨24 unchanged words⟩
⟨18 unchanged words⟩ its aliases, such as cs.LG for ai-ml. An exact multi-token phrase matches whole tokens. K17, PDAC, NCC, IIT, and LLM match the full names in the alias map. Hits are ordered by keyword overlap,then a domain match, then evidencestrength. That retrieval order stays ahead of the stored score. match_reason records that the stored score is evidence_strength. cross_domain also keeps claims from other domains that ⟨43 unchanged words⟩ evidence strength, a shortened excerpt, match_reason, source_url,whendoi,aandpaperpaper_idlinkwhenisthose are known, up to two supporting claims, and one contradicting claim when one is stored. Each related claim carries that same compact pointer. unpublished is true for a private unpublished paper. Full provenance stays on get_evidence. status_note on a consensus claim with no stored contradiction is: No stored contradiction; not independent multi-source consensus. evidence_note is: Extractor confidence in the claim statement, not study quality. integrity_warning is set when the statement unit looks truncated relative to the excerpt, or the magnitude looks extreme relative to the excerpt. The statement and the excerpt are not rewritten. Example: query_claims(query="transformer", domain="ai-ml" ⟨11 unchanged words⟩
⟨24 unchanged words⟩ "One stored contradicting claim, when one ispresent.present,claim_idwith the same compact source pointer."},"doi":{"default":"","description":"StoredstatementDOI,andthestrength.DataCite DOI of an arXiv id, or a bioRxiv paper id that is already a DOI. Empty when unknown. Full provenance stays on get_evidence.","title":"Doi","type":"string"},"domain":{"title":"Domain","type":"string"},"evidence_note":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Extractor confidence in the claim statement, not study quality. evidence_strength is that extractor confidence.","title":"Evidence Note"},"evidence_strength":{"title":"Evidence Strength","type":" ⟨8 unchanged words⟩ ":"Id","type":"string"},"integrity_warning":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Set when the statement unit looks truncated relative to the excerpt, or the magnitude looks extreme relative to the excerpt. The statement and the excerpt are not rewritten.","title":"Integrity Warning"},"match_reason":{"default":"","description":"Why this hit ranked here: keyword overlap, domain match, and
Find claims from a natural-language question.AThe central research concept and its scientific synonyms outrank incidental context words. Generic overlap that does not address the question ismatchedpenalized. A keyword with no function word stays on the lexical match. A question tries its contentphrases.phrases first. When that match isemptythin,thelongestextractedcontentconcepts,termsshorter phrases, and synonyms aresearchedcombinedasandkeywords.deduplicatedThosebefore ranking. Exact keywords are not required. Question hits stay onstrategyquestion,strategysetquestion. fallbacktois null when the content-phrase match already addressed the concept, keyword when every hit came from a literal concept or shorter phrase, anduseexpanded when a synonym supplied a hit. Rank puts thesameconcepthybridabove context words. Retrieval relevance stays ahead of evidence_strength, and match_reason records that the stored scoreasquery_claims:iskeywordevidence_strength.overlapA lexical phrase matches whole tokens. K17,thenPDAC,aNCC,domainIIT, and LLM match,thentheevidencefull names in the alias map.strength.notice is no_matching_claims when nothing matches. notice is no_evidence when every returned hit has an empty excerpt. Akeywordhitstayswithonan empty excerpt says no evidence in match_reason. exploratory true keeps matches under thelexicaldefaultmatch.relevance threshold. Each hit includesmatch_reason, source_url,whendoi,aandpaperpaper_idlinkwhenisthose are known, up to two supportingclaims, and one contradicting claim when one is stored. Each related claim carries that same compact pointer. unpublished is true for a private unpublished paper. Full provenancestays on get_evidence. status_note on a consensus claim with no stored contradiction is: No stored contradiction; not independent multi-source consensus. evidence_note is: Extractor confidence in the claim statement, not study quality. integrity_warning is set when the statement unit looks truncated relative to the excerpt, or the magnitude looks extreme relative to the excerpt. The statement and the excerpt are not rewritten. A domain keeps claims whose primary domain is ⟨44 unchanged words⟩ ", limit=5). Example: search_smart(natural_language="Whatlimitscauseshealthspan?",LLMdomain="longevityhallucinations in RAG systems?", limit=5).
⟨48 unchanged words⟩ ["ai-ml"],"title":"Domain"},"exploratory":{"default":false,"description":"Keep weaker matches that miss the central concept. Default false applies a light relevance threshold.","examples":[false],"title":"Exploratory","type":"boolean"},"limit":{"default":10,"description":"Page ⟨68 unchanged words⟩
⟨24 unchanged words⟩ "One stored contradicting claim, when one ispresent.present,claim_idwith the same compact source pointer."},"doi":{"default":"","description":"StoredstatementDOI,andthestrength.DataCite DOI of an arXiv id, or a bioRxiv paper id that is already a DOI. Empty when unknown. Full provenance stays on get_evidence.","title":"Doi","type":"string"},"domain":{"title":"Domain","type":"string"},"evidence_note":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Extractor confidence in the claim statement, not study quality. evidence_strength is that extractor confidence.","title":"Evidence Note"},"evidence_strength":{"title":"Evidence Strength","type":" ⟨8 unchanged words⟩ ":"Id","type":"string"},"integrity_warning":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Set when the statement unit looks truncated relative to the excerpt, or the magnitude looks extreme relative to the excerpt. The statement and the excerpt are not rewritten.","title":"Integrity Warning"},"match_reason":{"default":"","description":"Why this hit ranked here: keyword overlap, domain match, and
⟨41 unchanged words⟩ ["ai-ml"],"title":"Domain"},"flag_type":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Keep one flag_type. Example: lexical. Empty returns every type.","examples":["lexical"],"title":"Flag Type"},"limit":{"default":20,"description":"Page ⟨65 unchanged words⟩
⟨72 unchanged words⟩ "Created At","type":"string"},"cross_domain":{"default":false,"description":"True when the claims have different primary domains. Independent of flag_type.","title":"Cross Domain","type":"boolean"},"domain":{"title":"Domain","type": ⟨6 unchanged words⟩ ,"description":"Kind of link. lexical isshared-term context. cross_domain issharedevidenceterminologyoverlapwithacrossnoprimarysubstantivedomains.relation.contradictionscientific_connection is astoredconceptualdisagreement.or methodological relationship. innovation ishigher-strengthasynergynamed transfer oranextension.evidence-chaincontradictionextension.is a stored disagreement. Older rows may still say lexical_association,scientific_connectioncross_domain, or research_hypothesis."},"id":{"title" ⟨89 unchanged words⟩ be read. A missing value loads as lexical_association.\n\nlexical,cross_domain,andcontradiction,lexical_associationandare shared terminology with no substantive\nrelation. scientific_connection is a conceptual or methodological\nrelationship. innovationareistheakindstransferthe\nflagorpassextensionwrites.thatThecanthreebeoldernamed.\ncontradictionlabelsisstaya stored disagreement. cross_domain remains a validforolder\nlabel. New rowsalreadystorestored.a cross-domain link on cross_domain, not as the type.","enum":["lexical_association","scientific_connection"," ⟨59 unchanged words⟩
⟨83 unchanged words⟩ #/$defs/FlagType","default":"lexical_association","description":"lexical_associationKindisofexploratorylink. lexical is shared-term context.scientific_connectioncross_domainandisresearch_hypothesisevidenceareoverlapevaluatedacrossrelationships.primary domains. contradiction is a stored disagreement. innovation is higher-strength synergy or an evidence-chain extension. Older rows may still say lexical_association, scientific_connection, or research_hypothesis."},"id":{"title":"Id", ⟨5 unchanged words⟩ ,"type":"number"},"summary":{"description":"Short reason this pair was flagged. Use it to follow the link or skip it.","title":"Summary","type":"string"}} ⟨59 unchanged words⟩ should be read. A missing value loads aslexical_association.lexical_association.\n\nlexical, cross_domain, contradiction, and innovation are the kinds the\nflag pass writes. The three older labels stay valid for rows already stored.","enum":["lexical_association","scientific_connection","research_hypothesis","lexical","cross_domain","contradiction","innovation"],"title":"FlagType","type":" ⟨51 unchanged words⟩
⟨23 unchanged words⟩ Returns claim id, status, evidence strength, a shortened excerpt, and source_url when ashortenedpaperexcerpt.link is known. Example: get_updates(since="2026-01-01T00:00:00Z", ⟨14 unchanged words⟩
⟨12 unchanged words⟩ {"assertion_type":{"$ref":"#/$defs/AssertionType"},"contradicting":{"anyOf":[{"$ref":"#/$defs/RelatedClaim"},{"type":"null"}],"default":null,"description":"One stored contradicting claim, when one is present. claim_id, statement, and strength."},"domain":{"title":"Domain","type": ⟨14 unchanged words⟩ ":"Id","type":"string"},"match_reason":{"default":"","description":"Why this hit ranked here: keyword overlap, domain match, and evidence strength.","title":"Match Reason","type":"string"},"source_url":{"default":"","description":"Stable link to the paper when one is known. Empty when the claim has no pointer. Full provenance stays on get_evidence.","title":"Source Url","type":"string"},"statement":{"title":"Statement","type":"string"},"status":{"$ref":"#/$defs/ClaimStatus"},"supporting":{"description":"Up to two other claims from the same papers. Each has claim_id, statement, and strength.","items":{"$ref":"#/$defs/RelatedClaim"},"title":"Supporting","type":"array"}},"required":["id","statement","domain", ⟨42 unchanged words⟩ ,"returned","total"],"title":"Page","type":"object"},"RelatedClaim":{"description":"One claim shown beside a search hit.","properties":{"claim_id":{"title":"Claim Id","type":"string"},"source_url":{"default":"","description":"Stable link to that claim's paper when one is known. Empty when it has no pointer.","title":"Source Url","type":"string"},"statement":{"title":"Statement","type":"string"},"strength":{"title":"Strength","type":"number"}},"required":["claim_id","statement","strength"],"title":"RelatedClaim","type":"object"}},"properties": ⟨26 unchanged words⟩
⟨18 unchanged words⟩ its aliases, such as cs.LG for ai-ml.EvidenceHitsstrengthareordersorderedthosebyhits.keyword overlap, then a domain match, then evidence strength. cross_domain also keepsclaims from other domains that carry the domainas a secondary tag.When the query is one domain keyword and domain isomitted, claims inthat domain come first. Use this when you ⟨11 unchanged words⟩ Returns claim id, status, evidence strength,anda shortenedexcerpt.excerpt, match_reason, source_url when a paper link is known, up to two supporting claims, and one contradicting claim when one is stored. Full provenance stays on get_evidence. Example: query_claims(query="transformer", domain="ai-ml" ⟨11 unchanged words⟩
⟨12 unchanged words⟩ {"assertion_type":{"$ref":"#/$defs/AssertionType"},"contradicting":{"anyOf":[{"$ref":"#/$defs/RelatedClaim"},{"type":"null"}],"default":null,"description":"One stored contradicting claim, when one is present. claim_id, statement, and strength."},"domain":{"title":"Domain","type": ⟨14 unchanged words⟩ ":"Id","type":"string"},"match_reason":{"default":"","description":"Why this hit ranked here: keyword overlap, domain match, and evidence strength.","title":"Match Reason","type":"string"},"source_url":{"default":"","description":"Stable link to the paper when one is known. Empty when the claim has no pointer. Full provenance stays on get_evidence.","title":"Source Url","type":"string"},"statement":{"title":"Statement","type":"string"},"status":{"$ref":"#/$defs/ClaimStatus"},"supporting":{"description":"Up to two other claims from the same papers. Each has claim_id, statement, and strength.","items":{"$ref":"#/$defs/RelatedClaim"},"title":"Supporting","type":"array"}},"required":["id","statement","domain", ⟨42 unchanged words⟩ ,"returned","total"],"title":"Page","type":"object"},"RelatedClaim":{"description":"One claim shown beside a search hit.","properties":{"claim_id":{"title":"Claim Id","type":"string"},"source_url":{"default":"","description":"Stable link to that claim's paper when one is known. Empty when it has no pointer.","title":"Source Url","type":"string"},"statement":{"title":"Statement","type":"string"},"strength":{"title":"Strength","type":"number"}},"required":["claim_id","statement","strength"],"title":"RelatedClaim","type":"object"}},"properties": ⟨18 unchanged words⟩
⟨31 unchanged words⟩ strategy question, set fallback to keyword, andareuseorderedthebysamehowhybridmanyscorequestionastermsquery_claims:appearkeywordinoverlap,thethenstatementaordomainexcerpt.match, then evidence strength. A keyword stays onthe lexical match. Each hit includes match_reason, source_url when a paper link is known, up to two supporting claims, and one contradicting claim when one is stored. Full provenance stays on get_evidence. A domain keeps claims whose primary domain is ⟨56 unchanged words⟩
⟨12 unchanged words⟩ {"assertion_type":{"$ref":"#/$defs/AssertionType"},"contradicting":{"anyOf":[{"$ref":"#/$defs/RelatedClaim"},{"type":"null"}],"default":null,"description":"One stored contradicting claim, when one is present. claim_id, statement, and strength."},"domain":{"title":"Domain","type": ⟨14 unchanged words⟩ ":"Id","type":"string"},"match_reason":{"default":"","description":"Why this hit ranked here: keyword overlap, domain match, and evidence strength.","title":"Match Reason","type":"string"},"source_url":{"default":"","description":"Stable link to the paper when one is known. Empty when the claim has no pointer. Full provenance stays on get_evidence.","title":"Source Url","type":"string"},"statement":{"title":"Statement","type":"string"},"status":{"$ref":"#/$defs/ClaimStatus"},"supporting":{"description":"Up to two other claims from the same papers. Each has claim_id, statement, and strength.","items":{"$ref":"#/$defs/RelatedClaim"},"title":"Supporting","type":"array"}},"required":["id","statement","domain", ⟨42 unchanged words⟩ ,"returned","total"],"title":"Page","type":"object"},"RelatedClaim":{"description":"One claim shown beside a search hit.","properties":{"claim_id":{"title":"Claim Id","type":"string"},"source_url":{"default":"","description":"Stable link to that claim's paper when one is known. Empty when it has no pointer.","title":"Source Url","type":"string"},"statement":{"title":"Statement","type":"string"},"strength":{"title":"Strength","type":"number"}},"required":["claim_id","statement","strength"],"title":"RelatedClaim","type":"object"}},"description": ⟨70 unchanged words⟩
{"properties":{"cross_domain_only":{"default":false,"description":"When true, return only flags whose connected_domain is set. Default false.","examples":[false],"title":"Cross Domain Only","type":"boolean"},"domain":{"anyOf":[{"type":"string ⟨86 unchanged words⟩
⟨8 unchanged words⟩ "Claim Path","type":"array"},"claims":{"description":"Claims on this path, in claim_path order. Each item has claim_id, statement, and evidence strength.","items":{"$ref":"#/$defs/FlagPathClaim"},"title":"Claims","type":"array"},"connected_domain":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"Primary domain of the claim that does not match this flag's domain. Null when the claims share a primary domain.","title":"Connected Domain"},"created_at":{"format":"date-time","title": ⟨6 unchanged words⟩ ":"Domain","type":"string"},"flag_type":{"$ref":"#/$defs/FlagType","default":"lexical_association","description":"lexical_association is exploratory shared-term context. scientific_connection and research_hypothesis are evaluated relationships."},"id":{"title":"Id","type": ⟨18 unchanged words⟩ ],"title":"FlagHit","type":"object"},"FlagPathClaim":{"description":"One claim on a flag path. Evidence and contradictions stay on get_evidence.","properties":{"claim_id":{"title":"Claim Id","type":"string"},"statement":{"title":"Statement","type":"string"},"strength":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null,"title":"Strength"}},"required":["claim_id","statement"],"title":"FlagPathClaim","type":"object"},"FlagType":{"description":"How a flag should be read. A missing value loads as lexical_association.","enum":["lexical_association","scientific_connection","research_hypothesis"],"title":"FlagType","type":"string"},"Page":{"properties":{"limit" ⟨47 unchanged words⟩
Search claim statements andexcerptsexcerpts. A domain keeps claims whose primary domain is that value or one of its aliases,strongestsuchmatchesas cs.LG for ai-ml. Evidence strength orders those hits. cross_domain also keeps claims from other domains that carry the domain as a secondary tag. When the query is one domain keyword and domain is omitted, claims in that domain come first. Use this when you have keywords or ⟨35 unchanged words⟩
⟨105 unchanged words⟩ "type":"string"}},"properties":{"cross_domain":{"default":false,"description":"Also return claims from other primary domains that carry this domain as a secondary tag. Default false.","examples":[false],"title":"Cross Domain","type":"boolean"},"domain":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"RestrictPrimaryresultsdomaintooroneandomain.alias of it. Example: ai-ml.","examples":["ai-ml"] ⟨95 unchanged words⟩
Find claims from a natural-language question. AThisquestionbuildismatchesmatchedtextononly.its content phrases. When that match is empty, the longest content terms are searched as keywords. Those hits stay on strategy question, set fallback to keyword, and are ordered by how many question terms appear in the statement or excerpt. A keyword stays on the lexical match. A domain keeps claims whose primary domain is that value or one of its aliases. cross_domain also keeps claims from other domains that carry the domain as a secondary tag. Use this when the input is a question ⟨26 unchanged words⟩
{"properties":{"cross_domain":{"default":false,"description":"Also return claims from other primary domains that carry this domain as a secondary tag. Default false.","examples":[false],"title":"Cross Domain","type":"boolean"},"domain":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null,"description":"RestrictPrimaryresultsdomaintooroneandomain.alias of it. Example: ai-ml.","examples":["ai-ml"] ⟨75 unchanged words⟩
⟨94 unchanged words⟩ ,"type":"object"}},"description":"strategyquestionstaysmatches'lexical'contentuntilphrases.multi-hoplexicaltraversalis a keyword query.\n\nfallback isimplemented.keyword when the phrase match was empty and the claims come\nfrom a keyword search of the longest content terms.","properties":{"claims":{"items": ⟨3 unchanged words⟩ ":"Claims","type":"array"},"fallback":{"anyOf":[{"const":"keyword","type":"string"},{"type":"null"}],"default":null,"title":"Fallback"},"page":{"$ref":"#/$defs/Page"},"strategy":{"constdefault":"lexical","defaultenum":["lexical","question"],"title":"Strategy","type":"string" ⟨7 unchanged words⟩