[CLAUDE] Docs: S40 consolidate STATUS/HANDOFF 393KB→9KB + re-ground count drift + archive zero-loss

- STATUS.md 169.7KB→5.7KB, HANDOFF.md 223.6KB→3.4KB (giữ current+3 session, pointer logs)
- Full verbatim archived docs/_archive/{STATUS,HANDOFF}-preS40-fullhistory.md (zero info loss)
- Re-ground count: endpoints ~223→211, FE pages 53→65 (36+29), menu 85→~53; tables 84 confirmed correct (DbSet 77 + Identity 7)
- CLAUDE.md root patch: 33→40 mig, 60→84 bảng, 111→130 test
- S40 init audit: 7-agent live, RAG re-rank PASS 0.8789, 130 test PASS
- agent-memory S40 entries (investigator-codebase/api + test-specialist smoke audit)

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

.claude/agent-memory/investigator-api/MEMORY.md

@@ -28,7 +28,8 @@ Read-only EXTERNAL research specialist. WebFetch/WebSearch official docs + NuGet
 
 ## 📅 Recent activity (last 10 FIFO)
 
-- **2026-05-29 (S39 agent split setup):** NEW agent created từ split investigator. Seeded external-research half. Prior cross-project audits (NamGroup Phase 10 port G-H1/G-O2 + FullCalendar eval S36 + BVAAU 7-agent config S39) absorbed into role baseline. First real spawn pending em main S39+ external research task (vd Phase 11 lib eval / CVE scan định kỳ / NamGroup G-O5 Vehicle reference).
+- **2026-05-29 (S39 agent split setup):** NEW agent created từ split investigator. Seeded external-research half. Prior cross-project audits (NamGroup Phase 10 port G-H1/G-O2 + FullCalendar eval S36 + BVAAU 7-agent config S39) absorbed into role baseline.
+- **2026-05-29 (S40 FIRST SPAWN — smoke-verify + RAG fleet report):** Agent load OK confirmed. `list_projects` → 7 project, total **39,798 chunks**. Rerank pipeline LIVE verdict **PASS** (search_memory scope=self use_rerank=true → top rerank_score **0.8789**, 3 results all carry rerank_score). Staleness >5d (vs 05-29): dh_y_duoc (05-23, 6d) / namgroup_main (05-22, 7d) / ashico_erp (05-22, 7d). solution_erp 05-28 fresh-ish but missing S37-S39 content. `shared_global` = 0 chunks (chưa promote pattern nào). MINOR drift: namgroup_main actual 11306 (brief said 11305). vipix_ai_infra (1652) = AI_INFRA hub root `D:\...\AI_INFRA`. No re-ingest performed (report-only).
 
 ---
 

.claude/agent-memory/investigator-codebase/MEMORY.md

@@ -129,6 +129,8 @@ State machine 5 trạng thái phiếu PE: Nháp / Đã gửi duyệt / **Trả l
 
 ## 📅 Recent activity (last 10 FIFO)
 
+- **2026-05-29 (S40 STATE GROUNDING AUDIT — docs vs codebase drift):** Em spawn ~9K. **7 metric verify, 4 DRIFT / 3 khớp.** ✅ **Migrations=40** (path THẬT `src/Backend/SolutionErp.Infrastructure/Persistence/Migrations/*.cs` — STATUS glob `Migrations/*.cs` trả 0 SAI, skill `ef-core-migration` ghi đúng path nhưng stale "31 mig"). Mig cao nhất = `20260528090839_AddAttendances` (G-P1). ✅ **Gotchas=55** (format `### N.` NOT `#NN`, highest `### 55.`). ✅ **git clean** main. ❌ **DbSet=77** (docs 84, -7) — `grep "public DbSet<"` ApplicationDbContext.cs. ⚠️ **Endpoints=211** (docs ~223): 65 GET+69 POST+39 PUT+33 DELETE+5 PATCH. ❌ **FE pages fe-admin=36** (docs 53, -17 nặng; fe-user=29; 36+29=65≠53 nên không phải gộp). ❌ **Menu keys=53** (docs 85, -32 nặng — `MenuKeys.cs` const string=53 single-source; 85 nghi là seed row runtime sau inheritance expand NOT const định nghĩa). **Drift nặng nhất = Menu keys + FE pages** (over-count, docs trộn seed runtime vs const, hoặc S38 skeleton không sync ngược). **Surprises**: (a) Mig 37/38 THỰC = 37 `ExtendApplicableTypeForWorkflowApps` + 38 `AddProposals` TÁCH RIÊNG — KHÁC S37 memory tôi ghi "gộp 37+38 = Proposal" (recommend gộp KHÔNG được adopt). (b) DbSet 77<84 nghi vài entity Phase 10 (WorkflowApps/Attendances/Proposals/MeetingRooms G-* skeleton) chưa add DbSet hoặc gom chung — đáng audit sâu nếu skeleton wire thiếu. (c) Mig 40 thực = AddAttendances NOT documented "Mig 34-40 G-* port" range vẫn đúng. **Recommend**: sync STATUS.md 4 con số (Tables 77 / Endpoints 211 / FE pages 36 / Menu keys 53). Migration + gotcha + git là 3 con số TIN CẬY nhất. Token ~9K. Tag `[state-grounding, docs-drift, s40]`.
+
 - **2026-05-29 (cross-project — BVAAU 7-agent config extract):** Em đọc full 8 file `D:\Dropbox\CONG_VIEC\BENHVIEN_A_AU\SOURCE_CODDE\.claude\agents\` ~22K. **BVAAU split 4→7 trục research(2)/implement(2)/quality(3)**: investigator-api(blue READ external GetFly/Pancake docs) · investigator-codebase(cyan READ internal+sqlcmd) · implementer-domain(yellow WRITE `Domain/**`+`Application/**` only) · implementer-infra(orange WRITE `Infra/**`+`Api/**`+`Web/**`+`Web.Reports React`) · test-specialist(purple WRITE `tests/**` only NEW) · reviewer(red) · cicd-monitor(green). **Boundary key**: repo *interface*=domain, EF *config*=infra, test assertion=test-specialist (2 implementer KHÔNG viết test). **8-criteria REFUSE** cả 2 implementer (schema/UX/cross-stack>2layer/bug-reasoning/<30min/first-time/spec ambiguity>20%). **Token**: ~1.35M/heavy (~6.5× solo) optimized ~700K; Agent fresh ~150K cache-write / warm ~50K / SendMessage 5-30K/turn / parallel SendMessage ⚠️2.25-3×. **Tool diff CRITICAL**: cả 7 BVAAU agent có 5 RAG MCP (`search_memory+search_code+cross_project_search+store_memory+list_projects`) vs SOLUTION_ERP hiện chỉ 2 (`search_memory+cross_project_search`) → BVAAU +3 (`search_code` BM25 saves Read full file + `store_memory` instant index + `list_projects`). **Skill**: BVAAU `skills:` frontmatter KHÔNG có, total skill=0 (Phase 0 all TBD) — SOLUTION_ERP 6 skill+preload MẠNH HƠN, giữ nguyên khi port. **Curate rule**: >25KB archive `<YYYY-MM>.md`, >50KB hard limit. **Parallel patterns**: A=investigator-api→DTO→(infra+domain parallel)→test→reviewer→cicd; B=4 sync job song song file-disjoint. **Surprises**: (a) VIPIX guide `D:\...\VIPIX_MULTISITE_PROJECT\docs\guides\multi-agent-setup-guide.md` KHÔNG tồn tại (glob toàn project miss) → "16 spawn ~1.49M" claim không verify được, đừng cite. (b) 3 inconsistency BVAAU file: cicd ghi cả `.gitea/` (L25) lẫn `.github/` (L159) workflow path; investigator-codebase MEMORY path `/investigator/` mismatch tên agent; README "Setup checklist" L331 vẫn "4 sub-agent" chưa update 7. (c) BVAAU Phase 0 codebase RỖNG → config aspirational template chưa battle-test (vs SOLUTION_ERP 38 session proven). **Recommend port**: trục split + 8-criteria REFUSE + 5-category reviewer adversarial, GIỮ skill-frontmatter SOLUTION_ERP, ADD 3 RAG tool nếu MCP support. Token cost ~22K. Tag `[cross-project, bvaau-agent-config, port-pattern]`.
 
 - **2026-05-28 (S37 Plan G-O3 Đề xuất pre-flight 4 task):** Em spawn ~14K. **T1 Enum:** `ApprovalWorkflowApplicableType` ở `ApprovalWorkflow.cs:45-50` Mig 22 add — 3 value 1-based (DuyetNcc=1, DuyetNccPhuongAn=2, Contract=3) NOT 0-based. Mig 37 +1 value Proposal=4 (G-O3 first), reserve 5-8 cho G-O4..O6 future. **T2 PE pattern intel** (mirror skeleton Mig 38): main entity 25 field critical (Id/MaPhieu/Type/Phase 5-state/TenGoiThau/ProjectId/DepartmentId/DrafterUserId + ApprovalWorkflowId pin V2/CurrentStepIndex/CurrentLevelOrder + BudgetId/BudgetManual + RejectedFromPhase legacy + 8 nav list). LevelOpinion UNIQUE per-LEVEL NOT per-User (composite `(PEId, LevelId)`), FK Cascade Pe + Restrict Level, denorm SignedByFullName(200). CodeSequence **1 row per Prefix** (NOT N rows per year): `PE/YYYY/A` + `PE/YYYY/B` riêng, 3 col Prefix PK+LastSeq+UpdatedAt, tx SERIALIZABLE. ApproveV2Async logic 7 step (load eager → group Levels by Order = Cấp OR-of-N → match actor.Id ∈ ApproverUserId + Admin bypass → log Approval + UPSERT LevelOpinion auto sync → F2 skipToFinal check matchingLevel.AllowApproverSkipToFinal advance tới last Step+last Level Phase giữ ChoDuyet → advance normal levelOrder++ trong Step → next Step reset 1 → DaDuyet terminal). **T3 NamGroup Proposal**: 3 entity `TblRequest.cs` (25 cols generic + SoTien decimal + NguoiNhanTien + TaiKhoanNganHang advance payment) + `TblRequestConfig.cs` (CSV ListApproval inline + IsDirectManager) + `TblRequestListApproval.cs` (forward mechanism qua NguoiChuyenTiepId/NgayChuyenTiep + ChapThuanHoanToan int). **Verdict: SOL CLEAN-ROOM MẠNH HƠN** — NamGroup generic Request KHÔNG phân biệt Đề xuất/Yêu cầu/Phép (1 entity cover ALL). SOL dedicated `Proposal` entity + Workflow V2 pin FK > NamGroup inline CSV JSON. **T4 Mig 38 propose 4 bảng**: Proposals (Code `DX/YYYY/NNNN`, Title/Desc/Category nullable FK + free text fallback, AmountEstimate decimal? optional, Status ProposalStatus 5-state mirror PE, DepartmentId/DrafterUserId, ApprovalWorkflowId pin ApplicableType=4 + CurrentStep/Level pointer + SLA) + ProposalAttachments + ProposalLevelOpinions UNIQUE (PropId, LevelId) + ProposalCodeSequences 1 row Prefix `DX/YYYY`. **3 decision points em main chốt**: (1) CategoryId nullable FK + free text fallback (lesson Plan C drift 8 FK ZERO populated 1675 NV). (2) 5-state mirror PE NOT 7-state Contract (Đề xuất đơn giản hơn HĐ). (3) Dedicated `ProposalStatus` enum NOT reuse PurchaseEvaluationPhase (anti-pattern cross-module coupling). **Surprises**: (a) NamGroup TblRequest có `SoTien + NguoiNhanTien + TaiKhoanNganHang` advance payment — SOL có cần? Defer Mig 39 optional. (b) F1+F3+F4+F2 per-Level pattern Mig 29-31 — Proposal có nên mirror? Option B "đơn giản KHÔNG flag" tiết kiệm 3 chunk admin Designer. (c) **Recommend gộp Mig 37+38 → 1 Mig** `AddProposalsAndExtendApplicableType` (mirror Mig 20 Contract N-stage gộp). Token cost ~14K.
@@ -145,7 +147,6 @@ State machine 5 trạng thái phiếu PE: Nháp / Đã gửi duyệt / **Trả l
 
 - **2026-05-26 (S32 wrap — em main proxy update final state):** Session 32 đóng clean. Em chủ trì spawn em 2 lần S32: (1) startup verify ~10K (afaf6d52a6a59a844 alive) + (2) NamGroup audit pre-flight ~30K (a533c3e8ed4e03bfe hit limit 157K/72 tool_uses fallback em main solo audit). **Em main solo audit NamGroup directly** thay vì re-spawn: discovered NAMGROUP.Server\Data\Entities có 8+ TblNhanVien* (QtCongTac/QtDaoTao/QuanHeThanNhan/KyNangViTinh) + 4 bảng org chart (SoDoToChuc/SoDoKhoi/ChucDanh/ViTri) + Announcement/InternalDocument/TblMenu. **Plan G 11 module port DOCUMENTED migration-todos** với 4 quyết định chốt anh main (FULL 11 module + dbo single + Workflow V2 enum +5 + chunk per-module Plan riêng). **S33 priority spawn em pre-flight Plan G-H1 Hồ sơ NS** audit NamGroup TblNhanVien* 8 bảng + map fields → SOL EmployeeProfile schema (Mig 34 design 1 main + 5 satellite WorkHistory/Education/FamilyRelation/Skill/Document) — scope narrower 30 phút mục tiêu. **Curate Plan A3:** MEMORY archived 4 verbose entries q2 (S25 t1 5Q audit Bug Changelog + S26 Plan AG 5Q + S26 Plan AI RAG research + Plan B Contract V2 Q1-Q5 audit detail) → 27.7→19KB. **Pending tasks anh main S33 SendMessage gọi em:** (a) Plan G-H1 pre-flight NamGroup TblNhanVien* audit (PRIORITY HIGH first thing S33); (b) Plan B-Wrap test bundle pre-flight verify ContractWorkflowService ApproveV2Async test scenarios spec D-Bis trong migration-todos; (c) Phase 9 UAT smoke verify V2 contract create flow post-S29 deploy `QT-HD-V2-001` workflow. Token cost wrap ~5K. Tag: `[wrap, phase-9-to-phase-10, infra]`.
 
-- **2026-05-26 (S32 startup — em main proxy verify context + S31 RAG v1.3 baseline awareness):** S31 đóng ~1.5h ngày 2026-05-26 với RAG v1.3 baseline PASS recall@5=1.000 (11/11 queries) + avg_rerank=0.847. **Root cause S31 fix:** `AI_INFRA/claude-rag/lib/retrieval.py` xài API cũ `qdrant.search()` đã bị qdrant-client 1.18 xóa → đổi sang `query_points().points` API mới. **Gotcha #52 NEW** add vào `docs/gotchas.md` (Qdrant search removed). **Downstream impact:** MCP tools `mcp__rag-unified__search_memory` + `mcp__rag-unified__cross_project_search` live PASS post-CLI restart — test query "gotcha 52 qdrant search removed" top_k=3 trả 3 chunks rerank scores 0.515 / 0.479 / 0.461 (all dưới threshold 0.7 cao mong đợi vì query về symbol `qdrant.search` chỉ match doc rag-setup-plan.md historical, KHÔNG match gotcha #52 doc — doc chưa indexed lần re-index gần nhất hoặc chưa ingest gotchas folder). **Memory size 25.8 KB ĐÃ VƯỢT curate trigger 25KB** — em append entry này + flag em main curate cuối S32 (archive S25 + S26 verbose entries → `archive/2026-05-q2.md`). **Pending tasks anh có thể spawn em qua SendMessage:** (a) Plan B-Wrap test bundle BW1-BW7 (regression ApproveV2Async Contract V2 + ApplicableType=3 validate), (b) Phase 9 UAT audit Contract V2 wire prod usage (V1 7 contract + V2 sample `QT-HD-V2-001` smoke verify), (c) gotcha #52 doc verify trong RAG index (cross-check `docs/gotchas.md` đã re-ingest sau S31 fix chưa). **State delta S29 → S32:** 30 mig → 33 mig (Plan B Contract V2 Mig 32+33 + Plan CA role CatalogManager) · 59 → 60 tables · ~146 → ~148 endpoints · gotcha 47/48 → 52 (+4 #49 Plan B + #50 Plan CA INFRASTRUCTURE seed gate + #51 + #52 Qdrant) · 111 test PASS UNCHANGED (UAT defer test-after) · 23 → 25 memory user-level. Token cost spawn này ~10K.
 
 - **2026-05-22 (S29 wrap — Plan CA + Plan B pre-flight 2 spawns + 3 patterns NEW):** S29 close 2 big plans cumulative. **Plan CA (Move Cấu hình danh mục admin → eoffice)** 7 commits `06a441c..6eec8d7` push deployed Run #229+#230 PASS. **Plan B (Contract V2 wire mirror PE V2 pattern Mig 22-26)** 11 commits `58898e8..38f1c4d` push deployed Run #231+#232 PASS. **Investigator role 2 spawn:** (1) **Plan CA pre-flight audit** (agentId a7ab5576c77ee3730, ~15K token): map 9 menu danh mục terrain + GOTCHA tree-inherit Catalogs↔Master + recommend chunk structure A/B/C/D. fe-user component parity verified (DataTable + PageHeader + PermissionGuard + 6 shadcn ui). DbInitializer SeedAdminPermissionsAsync admin role grant CRUD 27 menu key. Permission Matrix dynamic load /api/menus → tự reflect post-move. (2) **Plan B pre-flight audit Contract V1 state** (agentId abf91b30391fb0cdb, ~25K token): map `Contract.cs` 25 fields (V1 only) + `ContractPhase` enum 12 values (TraLai=98 + TuChoi=99 ready) + `ContractWorkflowService.cs` 220 LOC V1 only + sqlcmd prod 7 V1 contract (Phase 2-8 active workflow) + PE Mig 22-26 reference templates + impact assessment (COEXIST V1+V2, KHÔNG drop V1) + re-chunk 6 chunks (split A → A1 entity + A2 mig). 3 surprise: ApprovalWorkflows Prod ZERO ApplicableType=3 (Chunk A2 must seed sample) + `RejectedAtStepIndex` drift deprecated + ContractType 7 variants generic ApplicableType=3. **Patterns proven NEW Investigator perspective S29:** (a) **9-menu permission terrain map pattern** (Plan CA) — Investigator inventory file paths + line ranges + component parity check + recommend chunk before delegate Implementer. ROI cao. (b) **V1+V2 coexist boundary** (Plan B) — KHÔNG drop V1 vì 7 prod contract pinned. Mirror PE Mig 22-24 spec. (c) **Reference template paths cross-module mirror** (PE → Contract V2): Investigator cite EXACT line ranges (PE Mig 23:14-31, Mig 24:14-17, Mig 26 UNIQUE composite, Service ApproveV2Async 446-634, 519-546 UPSERT, 774-783 ResolveActor) → em main + Implementer chỉ cần mirror rename entity. ROI rất cao saving 30%+ time. **Anti-patterns observed:** (a) Investigator KHÔNG verify endpoint exists end-to-end qua live curl — defer to CICD Monitor stage 4. (b) Investigator output sometimes verbose >700 words → em main hỏi cô đọng <500 next time.
 

.claude/agent-memory/test-specialist/MEMORY.md

@@ -50,6 +50,7 @@ Test theo CODE (single source truth), document mismatch header comment + report.
 ## 📅 Recent activity (last 10 FIFO)
 
 - **2026-05-29 (S39 agent split setup):** NEW dedicated agent. Seeded test patterns (10 reflection authz + 11 infra helper + 12 InternalsVisibleTo + #48 SQLite tie-break + spec drift S34). Inherited coverage gap backlog 4 priority items từ S36 Reviewer audit (130 PASS baseline). First spawn pending em main S39+ test bundle task (recommend Gap 1 Holiday composite UNIQUE first).
+- **2026-05-29 (S40 baseline audit smoke):** CONFIRMED 130 PASS (Domain 58 + Infra 72), 0 fail/skip, ~15s. Runner count authoritative; raw `[Fact]/[Theory]` attr = 48+70 (Theory→InlineData expand). Infra spread 15 files. Gap re-verified vs prod: EmployeesController+HrmConfigsController EXIST, authz regression chỉ ApprovalWorkflowsV2Controller (gotcha #44 gap real). Proposal = Domain entity + EF config only, CHƯA có ApproveV2Async service (S37 skeleton, defer đúng). Agent load OK. AUDIT-only, no write.
 
 ---