Unverified50% confidenceFactExact time
M|BOX's index schema classifies single-word attributes (such as first name, last name, city) as term fields, multi-word names (such as company) as phrase fields, and customer ID and email as non-searchable fields.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Building Millisecond-Level Identity Resolution for AI Agents with M|BOX and Python
youtubeOrangeBrackets10/27/2026
Related Entities
Related Claims
UnverifiedEntity-driven retrieval refines the matching granularity from 'entire sentence to entire name' to 'term to entity': even if a table name is an internal abbreviation, as long as that abbreviation has co-occurred with a certain business term in the query logs, a mapping can be established.62% similarUnverifiedLlamaIndex将文档切分为Node并建立多种索引结构，包括向量索引（VectorStoreIndex）、关键词索引（KeywordTableIndex）、知识图谱索引（KnowledgeGraphIndex），检索时由Retriever召回相关Node，Response Synthesizer组合内容送入LLM生成答案60% similarUnverifiedThe difference in keyword separators stems from each platform's search index parsing logic: Adobe Stock treats spaces within a quoted cell as keyword boundaries, while Vecteezy and Dreamstime split on commas.60% similarUnverified`vercel connect discover` supports searching by name, alias, keyword, and description.59% similarUnverifiedAllowlists should be written as exact hostnames rather than wildcard suffixes, because suffix matching such as *.example.net would include pre-release subdomains as well.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1107112API
curl https://kongchang.com/api/v1/knowledge/claims/1107112MCP
get_claim(id=1107112)