MerCry
/
ai-robot-core
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[AC-AISVC-02, AC-AISVC-16] 多个需求合并 #1

Merged
MerCry merged 45 commits from feature/prompt-unification-and-logging into main 2026-02-25 17:17:35 +00:00
Conversation 0 Commits 45 Files Changed 136 +28460 -1
MerCry commented 2026-02-25 17:16:59 +00:00
Owner
Copy Link
No description provided.
MerCry added 45 commits 2026-02-25 17:17:00 +00:00
0a167d69f0 feat(ai-service): implement LLM Adapter for T3.1 [AC-AISVC-02, AC-AISVC-06] 
- Add LLMClient abstract base class with generate/stream_generate interfaces
- Implement OpenAIClient with httpx for OpenAI-compatible API calls
- Add retry logic with tenacity for timeout handling
- Support both non-streaming and streaming generation
- Add comprehensive unit tests for LLM Adapter
- Fix entities.py JSON column type for SQLModel compatibility
550d0d8498 feat(ai-service): implement context merging for T3.2 [AC-AISVC-14, AC-AISVC-15] 
- Add ContextMerger class for combining local and external history
- Implement message fingerprint computation (SHA256 hash)
- Implement deduplication: local history takes priority
- Implement token-based truncation using tiktoken
- Add comprehensive unit tests (20 test cases)
c9f2c1eb3a feat(ai-service): implement SSE event generator for message events [AC-AISVC-07] 
- Optimize Orchestrator streaming output with LLM client integration
- Implement _stream_from_llm() to wrap LLM chunks as message events
- Implement _stream_mock_response() for demo/testing
- Add SSE event format tests (message/final/error)
- Fix by_alias=True for final event JSON output
- 79 tests passing
66fa2d2677 feat(ai-service): implement confidence calculation for T3.3 [AC-AISVC-17, AC-AISVC-18, AC-AISVC-19] 
- Add ConfidenceCalculator class for confidence scoring
- Implement retrieval insufficiency detection (hit count, score threshold, evidence tokens)
- Implement confidence calculation based on retrieval scores
- Implement shouldTransfer logic with configurable threshold
- Add transferReason for low confidence scenarios
- Add comprehensive unit tests (19 test cases)
- Update config with confidence-related settings
00953196c9 feat(ai-service): implement SSE state machine and error handling [AC-AISVC-08, AC-AISVC-09] 
- Integrate state machine in API layer for proper event sequencing
- Ensure message* -> final/error -> close sequence
- Prevent events after final/error
- Convert streaming exceptions to error events
- Add comprehensive state machine tests
- 117 tests passing
6d54030e0d feat(ai-service): implement complete Orchestrator generation pipeline for T3.4 [AC-AISVC-01, AC-AISVC-02] 
- Integrate Memory, ContextMerger, Retriever, LLMClient, ConfidenceCalculator
- Implement 8-step generation pipeline:
  1. Load local history from Memory
  2. Merge with external history (dedup + truncate)
  3. RAG retrieval (optional)
  4. Build prompt with context and evidence
  5. LLM generation
  6. Calculate confidence
  7. Save messages to Memory
  8. Return ChatResponse
- Add GenerationContext dataclass for tracking intermediate results
- Implement fallback response mechanism for error handling
- Add 21 unit tests for OrchestratorService
- All 138 tests passing
e4dc3d97e2 feat(ai-service): add Phase 5 integration and contract tests [AC-AISVC-10,11,17,18] 
- Add multi-tenant integration tests (test_integration_tenant.py)
- Add RAG smoke tests (test_rag_smoke.py)
- Add API contract tests (test_contract.py)
- 184 tests passing
- Phase 4 & 5 complete
43ce9837a1 feat(ai-service): add contract validation tests for T3.5 [AC-AISVC-02] 
- Verify ChatResponse fields match OpenAPI contract
- Test required fields: reply, confidence, shouldTransfer
- Test optional fields: transferReason, metadata
- Test JSON serialization uses camelCase (by_alias=True)
- Test confidence range validation [0.0, 1.0]
- Test ChatRequest contract compliance
- Test ErrorResponse contract compliance
- Test SSEFinalEvent matches ChatResponse structure
- Test SSEErrorEvent matches ErrorResponse structure
- Add end-to-end contract validation with OrchestratorService
- All 184 tests passing
71e966d62a docs(ai-service): update progress files [AISVC-ALL] all phases completed 
- Update spec/ai-service/progress.md with Phase 3 completion details
- Update spec/ai-service/tasks.md status to completed
- All 5 phases (23 tasks) completed: T1.1-T1.6, T2.1-T2.5, T3.1-T3.5, T4.1-T4.4, T5.1-T5.3
- 184 tests passing
- All AC markers covered: AC-AISVC-01~AC-AISVC-20
fb3c8ca2aa feat(ai-service): v0.2.0 前后端联调真实对接 
实现内容:
- 新增知识库实体模型 (KnowledgeBase, Document, IndexJob)
- 新增 KBService 服务层，支持文档上传、存储、索引任务管理
- 实现知识库管理 API 真实对接 (POST/GET /admin/kb/documents)
- 实现索引任务状态查询 API (GET /admin/kb/index/jobs/{jobId})
- 实现 RAG 实验室真实向量检索 (POST /admin/rag/experiments/run)
- 实现会话监控真实数据库查询 (GET /admin/sessions)

规范更新:
- requirements.md: v0.1.0 -> v0.2.0, 新增 AC-AISVC-21~28
- tasks.md: v0.1.0 -> v0.2.0, 新增 Phase 6 (9个任务)
- openapi.admin.yaml: L0 -> L1, 更新 x-requirements 映射

验收标准: AC-AISVC-21, AC-AISVC-22, AC-AISVC-23, AC-AISVC-24,
          AC-AISVC-25, AC-AISVC-26, AC-AISVC-27, AC-AISVC-28
5148c6ef42 feat(AISVC-T6.9): 前后端接口对接修正及Dashboard/RAG Lab功能完善 
## 后端修改
- 新增 Dashboard 统计 API (/admin/dashboard/stats)
- 新增知识库列表 API (/admin/kb/knowledge-bases)，返回文档数量
- 会话列表 API 新增 tenantId 字段
- KBService 新增 list_knowledge_bases 方法

## 前端修改
- Dashboard 页面对接真实后端 API
- RAG Lab 知识库选择器显示文档数量
- Monitoring 页面修复数据映射
- 新增 dashboard.ts API 文件
- kb.ts 新增 listKnowledgeBases 函数
4b64a4dbf4 feat(AISVC-T6.9): 集成Ollama嵌入模型修复RAG检索问题 
## 问题修复
- 替换假嵌入(SHA256 hash)为真实Ollama nomic-embed-text嵌入
- 修复Qdrant客户端版本不兼容导致score_threshold参数失效
- 降低默认分数阈值从0.7到0.3

## 新增文件
- ai-service/app/services/embedding/ollama_embedding.py

## 修改文件
- ai-service/app/api/admin/kb.py: 索引任务使用真实嵌入
- ai-service/app/core/config.py: 新增Ollama配置,向量维度改为768
- ai-service/app/core/qdrant_client.py: 移除score_threshold参数
- ai-service/app/services/retrieval/vector_retriever.py: 使用Ollama嵌入
c1d76093aa feat(AISVC-T7): 嵌入模型可插拔设计与文档解析支持 [AC-AISVC-29, AC-AISVC-30, AC-AISVC-31, AC-AISVC-32, AC-AISVC-33, AC-AISVC-34, AC-AISVC-35, AC-AISVC-36, AC-AISVC-37, AC-AISVC-38, AC-AISVC-39, AC-AISVC-40, AC-AISVC-41] 
- 新增 EmbeddingProvider 抽象基类和工厂模式 [AC-AISVC-29, AC-AISVC-30]
- 实现 OllamaEmbeddingProvider 和 OpenAIEmbeddingProvider [AC-AISVC-29, AC-AISVC-30]
- 新增 EmbeddingConfigManager 支持配置热更新 [AC-AISVC-31, AC-AISVC-32]
- 新增 DocumentParser 抽象接口和工厂类 [AC-AISVC-33]
- 实现 PDF/Word/Excel/Text 文档解析器 [AC-AISVC-33, AC-AISVC-34, AC-AISVC-35]
- 新增嵌入管理 API 端点 [AC-AISVC-38, AC-AISVC-39, AC-AISVC-40, AC-AISVC-41]
- 更新文档上传流程支持多格式文档解析 [AC-AISVC-36, AC-AISVC-37]
- 更新 OpenAPI 契约添加嵌入管理接口
- 添加数据库初始化脚本
- 更新规范文档标记 Phase 7 完成
e9fee2f80e feat(AISVC): Excel/CSV parser converts to JSON format for better RAG retrieval [AC-AISVC-35] 
- ExcelParser: convert spreadsheet data to JSON records with header as keys
- CSVParser: convert CSV data to JSON records preserving structure
- Add _sheet field to identify worksheet source in Excel output
- Preserve numeric types (int/float/bool) in JSON output
- Support UTF-8 and GBK encoding fallback for CSV files
fc53fdc6ac feat(AISVC-T8): LLM配置管理与RAG调试输出支持 [AC-AISVC-42, AC-AISVC-43, AC-AISVC-44, AC-AISVC-45, AC-AISVC-46, AC-AISVC-47, AC-AISVC-48, AC-AISVC-49, AC-AISVC-50] 
- 新增 LLMProviderFactory 工厂类支持 OpenAI/Ollama/Azure [AC-AISVC-42]
- 新增 LLMConfigManager 支持配置热更新 [AC-AISVC-43, AC-AISVC-44]
- 新增 LLM 管理 API 端点 [AC-AISVC-42~AC-AISVC-46]
- 更新 RAG 实验接口支持 AI 回复生成 [AC-AISVC-47, AC-AISVC-49]
- 新增 RAG 实验流式输出 SSE [AC-AISVC-48]
- 支持指定 LLM 提供者 [AC-AISVC-50]
- 更新 OpenAPI 契约添加 LLM 管理接口
- 更新前后端规范文档 v0.4.0 迭代
4579159c0a feat(ui): 前端浅色调风格优化与下拉框显示修复 
- 建立全局浅色调样式系统，统一配色风格
- 优化导航栏设计，添加品牌标识
- 修复下拉框被遮挡问题，添加 teleported 和 popper 配置
- 优化 LLM 选择器中当前配置标签的显示
- 重构控制台页面，采用白色卡片风格
- 统一所有页面的视觉风格，提升用户体验
08e84d194f feat: RAG实验室参数持久化与LLM选择器优化 [AC-ASA-05, AC-ASA-14, AC-ASA-15] 
- 新增ragLab store，使用localStorage持久化RAG实验参数
- 修复LLM选择器placeholder显示逻辑，正确显示当前配置
- 添加'已选择'标签区分用户选择与系统默认配置
- 优化下拉框样式防止标签被遮挡
10591ea8fd feat: 优化RAG客服回复prompt模板 [AC-ASA-05, AC-ASA-19] 
- 重构客服角色设定，增加系统指令部分
- 添加口语化、亲切语气等7项客服准则
- 优化知识库内容引用格式
- 移除暴露AI身份的提示词
eb45629b67 fix: LLM选择器当前配置标签显示逻辑 [AC-ASA-14, AC-ASA-15] 
- 修复placeholder动态显示当前配置提供者名称
- 添加computedPlaceholder计算属性
- 优化标签显示逻辑，区分'当前配置'和'已选择'
PR Check (SDD Full Gate) / sdd-full-gate (pull_request) Successful in 2s Details
f631f1dea0 feat: 统一提示词模板并添加全量提示词日志 [AC-AISVC-02, AC-ASA-19, AC-ASA-20] 
- 新增 prompts.py 集中管理系统提示词和证据格式化
- orchestrator.py 添加全量提示词日志打印
- openai_client.py 添加全量提示词日志打印（支持普通和流式）
- rag.py 重构使用统一的提示词模板
MerCry merged commit 1e3fe808e8 into main 2026-02-25 17:17:35 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels
No items
No Label
Milestone
Clear milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
No Assignees
1 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: MerCry/ai-robot-core#1
No description provided.
Delete Branch "feature/prompt-unification-and-logging"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?