Unverified50% confidenceSolutionExact time
The project adopts a layered directory structure of ingest → vector_store → sparse_retriever → fusion → reranker → pipeline → generate → eval, decoupling the ingestion side from the query side.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/5/2026
First Seen
Valid until: 1/3/2027
Sources
Related Entities
Related Claims
UnverifiedThe "query–analyze–generate–distribute" pipeline involves multiple coordinating modules, including a data ingestion layer, an LLM orchestration layer, a tool execution layer, and an output distribution layer.66% similarUnverified命令行工具应遵循目标平台目录约定、区分配置数据缓存三类文件、提供环境变量覆盖能力并使用成熟跨平台库处理路径65% similarUnverifiedRAG管线的关键组件包括文档解析与分块、向量嵌入、向量数据库存储（Milvus、Pinecone、Weaviate等）、检索策略（混合检索、重排序）以及上下文注入与Prompt组装62% similarUnverifiedAnthropic设计了CLAUDE.md的分层配置体系，支持项目根目录、子目录和用户主目录三个层级61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/976377API
curl https://kongchang.com/api/v1/knowledge/claims/976377MCP
get_claim(id=976377)