Files

HaHafeng 40c2f8e148 feat(rag): Complete RAG engine implementation with pgvector

Major Features:
- Created ekb_schema (13th schema) with 3 tables: KB/Document/Chunk
- Implemented EmbeddingService (text-embedding-v4, 1024-dim vectors)
- Implemented ChunkService (smart Markdown chunking)
- Implemented VectorSearchService (multi-query + hybrid search)
- Implemented RerankService (qwen3-rerank)
- Integrated DeepSeek V3 QueryRewriter for cross-language search
- Python service: Added pymupdf4llm for PDF-to-Markdown conversion
- PKB: Dual-mode adapter (pgvector/dify/hybrid)

Architecture:
- Brain-Hand Model: Business layer (DeepSeek) + Engine layer (pgvector)
- Cross-language support: Chinese query matches English documents
- Small Embedding (1024) + Strong Reranker strategy

Performance:
- End-to-end latency: 2.5s
- Cost per query: 0.0025 RMB
- Accuracy improvement: +20.5% (cross-language)

Tests:
- test-embedding-service.ts: Vector embedding verified
- test-rag-e2e.ts: Full pipeline tested
- test-rerank.ts: Rerank quality validated
- test-query-rewrite.ts: Cross-language search verified
- test-pdf-ingest.ts: Real PDF document tested (Dongen 2003.pdf)

Documentation:
- Added 05-RAG-Engine-User-Guide.md
- Added 02-Document-Processing-User-Guide.md
- Updated system status documentation

Status: Production ready

2026-01-21 20:24:29 +08:00

01-整体开发计划

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

02-里程碑规划

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

03-每周计划

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

04-技术决策

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

05-技术债务

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

01-文档处理引擎设计方案_v1.2.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

01-知识库引擎架构设计_v1.2.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

08-技术方案-跨语言检索优化.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

2026-01-11-数据库事故总结.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB前端问题修复报告.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB前端验证指南.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB功能审查报告-阶段0.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB和RVW功能迁移计划.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB精细化优化报告.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB迁移-超级安全执行计划.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB迁移-阶段1完成报告.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB迁移-阶段2完成报告.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB迁移-阶段2进行中.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB迁移-阶段3完成报告.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

PKB迁移-阶段4完成报告.md

feat(rag): Complete RAG engine implementation with pgvector

2026-01-21 20:24:29 +08:00

README.md

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

V2.2版本变化说明.md

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

下一阶段行动计划-V2.0-模块化架构优先.md

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

下一阶段行动计划-V2.1-务实版.md

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

下一阶段行动计划-V2.2-前端架构优先版.md

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

下一阶段行动计划-V2.2-完整版.md

feat(admin): Add user management and upgrade to module permission system

2026-01-16 13:42:10 +08:00

README.md

项目管理

用途： 项目进度跟踪、计划管理、决策记录
维护者： 项目管理组
最后更新： 2025-11-07

📋 目录结构

08-项目管理/
├── 01-整体开发计划/        # 长期规划（季度/年度）
│   └── README.md
├── 02-里程碑规划/           # 中期规划（周/月）
│   ├── README.md
│   ├── 04-里程碑4-架构改造.md
│   └── 05-里程碑5-ASL模块开发.md
├── 03-每周计划/             # 短期跟踪（日/周）
│   ├── README.md
│   └── 周次-进度报告模板.md
├── 04-决策记录/             # 重要决策归档（ADR）
│   ├── README.md
│   ├── ADR模板.md
│   └── ADR-001-Schema隔离暂缓执行.md
└── 下一阶段行动计划-ASL启动.md  # 当前阶段行动计划 ⭐

🎯 当前阶段

下一阶段行动计划（2025-11-07 至 2025-12-06）

核心目标： Schema隔离 + 前端统一架构 + ASL核心功能

四周规划：

Week 1： Schema隔离（3详细+7空）
Week 2： 前端统一架构 + 后端分层 ⭐⭐⭐
Week 3-4： ASL核心功能（4个LLM筛选）
Week 5+： LLM网关统一 + 其他模块

当前版本： 下一阶段行动计划-V2.2-前端架构优先版.md ⭐⭐⭐

版本历史：

V2.2版（前端架构优先） - 2025-11-09 ⭐ 当前版本
- 核心调整：前端架构提前到Week 2，避免返工
- LLM网关推迟到Week 5，聚焦核心
- 4个LLM已配置（DeepSeek/GPT-5/Claude-4.5/Qwen）
V2.1版（务实版） - 2025-11-09
- 核心调整：3详细+7空Schema策略
- Just-in-time设计，避免过度设计
V2.0版（模块化架构优先） - 2025-11-07
- 核心调整：Schema隔离 + 独立部署架构
V1.0版（ASL启动） - 2025-11-06
- 初始版本：直接启动ASL开发

🚀 快速导航

我想...

📋 了解项目整体规划
→ 查看整体开发计划

🎯 了解当前阶段要做什么
→ 查看下一阶段行动计划-V2.2（前端架构优先版） ⭐⭐⭐

📊 查看本周要做什么
→ 查看每周计划

🔍 了解某个决策为什么这样做
→ 查看决策记录

维护者： 项目管理组
创建日期： 2025-11-06
最后更新： 2025-11-07

README.md Unescape Escape

项目管理

📋 目录结构

🎯 当前阶段

下一阶段行动计划（2025-11-07 至 2025-12-06）

🚀 快速导航

我想...

README.md