HaHafeng
3d35e9c58b
Major Updates: - Add StreamingService with OpenAI Compatible format (backend/common/streaming) - Upgrade Chat component V2 with Ant Design X integration - Implement AIA module with 12 intelligent agents - Create AgentHub with 100% prototype V11 restoration - Create ChatWorkspace with streaming response support - Add ThinkingBlock for deep thinking display - Add useAIStream Hook for OpenAI Compatible stream handling Backend Common Capabilities (~400 lines): - OpenAIStreamAdapter: SSE adapter with OpenAI format - StreamingService: unified streaming service - Support content and reasoning_content dual streams - Deep thinking tag processing (<think>...</think>) Frontend Common Capabilities (~2000 lines): - AIStreamChat: modern streaming chat component - ThinkingBlock: collapsible deep thinking display - ConversationList: conversation management with grouping - useAIStream: OpenAI Compatible stream handler Hook - useConversations: conversation state management Hook - Modern design styles (Ultramodern theme) AIA Module Frontend (~1500 lines): - AgentHub: 12 agent cards with timeline design - ChatWorkspace: fullscreen immersive chat interface - AgentCard: theme-colored cards (blue/yellow/teal/purple) - 5 phases, 12 agents configuration - Responsive layout (desktop + mobile) AIA Module Backend (~900 lines): - agentService: 12 agents config with system prompts - conversationService: refactored with StreamingService - attachmentService: file upload skeleton (30k token limit) - 12 API endpoints with authentication - Full CRUD for conversations and messages Documentation: - AIA module status and development guide - Universal capabilities catalog (11 services) - Quick reference card for developers - System overview updates Testing: - Stream response verified (HTTP 200) - Authentication working correctly - Auto conversation creation working - Deep thinking display working - Message input and send working Status: Core features completed (85%), attachment and history loading pending
通用能力层
层级定义： 跨业务模块共享的核心技术能力
核心原则： 可复用、高内聚、独立部署
📋 能力清单
| 能力 | 说明 | 复用率 | 优先级 | 状态 |
|---|---|---|---|---|
| 01-LLM大模型网关 | 统一管理LLM调用、成本控制、模型切换 | 71% (5/7) | P0 | ⏳ 待实现 |
| 02-文档处理引擎 | PDF/Docx/Txt提取、OCR、表格提取 | 86% (6/7) | P0 | ✅ 已实现 |
| 03-RAG引擎 | 向量检索、语义搜索、RAG问答 | 43% (3/7) | P1 | ✅ 已实现 |
| 04-数据ETL引擎 | Excel JOIN、数据清洗、数据转换 | 29% (2/7) | P2 | ⏳ 待实现 |
| 05-医学NLP引擎 | 医学实体识别、术语标准化 | 14% (1/7) | P2 | ⏳ 待实现 |
🎯 设计原则
1. 可复用性
- 多个业务模块共享
- 避免重复开发
2. 独立部署
- 可以独立为微服务
- 支持独立扩展
3. 高内聚
- 每个能力职责单一
- 接口清晰
4. 领域知识
- 包含业务领域知识
- 不是纯技术组件
📊 复用率分析
LLM网关 - 71%复用率（最高优先级）
- AIA（AI智能问答）
- ASL（AI智能文献）
- PKB（个人知识库）
- DC（数据清洗）
- RVW（稿件审查）
文档处理引擎 - 86%复用率（已实现）
- ASL、PKB、DC、SSA、ST、RVW
RAG引擎 - 43%复用率（已实现）
- AIA、ASL、PKB
📚 快速导航
快速上下文
- [AI对接] 通用能力快速上下文.md - 2-3分钟了解通用能力层
核心能力
🔗 相关文档
最后更新： 2025-11-06
维护人： 技术架构师