Files
solution-erp/.claude/agent-memory/investigator-codebase/archive/2026-05-q4.md
pqhuy1987 f36aab8934
All checks were successful
Deploy SOLUTION_ERP / build-deploy (push) Successful in 4m52s
[CLAUDE] Docs: adopt Harness-9 — L2 archive dark-matter recovery (4 sub) + adap 2-workflow mandate (S70)
3-stage Workflow run-id evidence: investigate wf_be952f3c-97f / implement wf_a58e0d15-beb / audit wf_9520d8cd-4fe.

PART 1 (L2 recovery): 4 over-cap sub (cicd-monitor/investigator-codebase/reviewer/implementer-backend)
curated L1->L2 byte-exact + archive/_INDEX.md (substring sha-keyed pointers, no line-hints)
+ <period>.gist.md (4-field distill, distill-gen:1, verbatim frozen). All 4 MEMORY.md now < 25KB
auto-inject cap (closes P1 curate-debt). ~240KB archive no longer RAG-dark. 0-byte-loss git+sha
verified (Stage C audit + em-main self-gate on 2 reviewer StructuredOutput no-returns). Read-side
gap fixed (MEMORY.md L5 header -> _INDEX). + memory-budget.json (seed-by-measure) +
scripts/measure-agent-memory.ps1 + .ragignore guard.

PART 2/3 (process mandate): every adap = 2 separate workflows (implement + review) + report with
run-id; short-but-needs-confirm still requires review. Codified in .claude/commands/adap-apply.md
+ agents/README.md (Upgrade S70) + session-start.md (§2.1.2 budget-audit, pending-restart).

adap-report + email-back to AI_INFRA (body-hash 7c07b716e775).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-17 23:52:51 +07:00

2.7 KiB
Raw Blame History

Investigator-Codebase — Archive Q4 2026-05 (S39-S40 era)

Moved from MEMORY.md L1 (Harness-9 curate 2026-06-17). Verbatim — byte-exact, no reflow. Created S69: MEMORY.md referenced archive/2026-05-q4.md (S40 curate stub + footer) but the file was absent on disk (content was git-only d2f52ba). This file now holds the 3 May FIFO-overflow entries; the S29-S37 batch remains in git d2f52ba.


  • 2026-05-29 (S39 BVAAU 7-agent extract): Đọc 8 file BVAAU .claude/agents/ ~22K. Split 4→7 trục research(2)/implement(2)/quality(3). Boundary: repo interface=domain, EF config=infra, test=test-specialist. Tool: cả 7 agent 5 RAG MCP (+search_code BM25 +store_memory +list_projects). BVAAU Phase 0 codebase RỖNG → aspirational template chưa battle-test; SOLUTION_ERP giữ 6 skill + backend/frontend split (thay domain/infra cho 2-FE fit). VIPIX guide claim KHÔNG verify được (file miss). Tag [cross-project, bvaau, port].

  • 2026-05-29 (S40 STATE GROUNDING): 7 metric verify. Migrations=40 (path .../Persistence/Migrations/*.cs, last AddAttendances). Gotchas=55 (### N.). git clean. DbSet=77 nhưng SQL tables=84 (em main verify .ToTable() ModelSnapshot — 77+7 Identity, "84 docs ĐÚNG", DbSet count sai 7). Endpoints=211 (docs ~223). FE pages fe-admin 36+fe-user 29=65 (docs 53 under-count). Menu keys=53 const (docs 85 over-count). 3 số tin cậy nhất = mig/gotcha/git. Lesson: tables phải count ToTable KHÔNG DbSet. Tag [state-grounding, docs-drift, s40].

  • 2026-05-30 (P11-A WorkflowApps wire pre-flight): 4 module Leave/OT/Travel/Vehicle. Schema pin ĐÃ CÓ SẴN (Mig 39): Office/{Module}.cs đều có ApprovalWorkflowId?+CurrentApprovalLevelOrder?+WorkflowAppStatus (5-state khớp ProposalStatus). SKELETON tại Application/Office/WorkflowAppsFeatures.cs:11-15 (chỉ Create+List, KHÔNG Approve/Reject/Return/GetById). Proposal = mirror HOÀN HẢO (cùng Office ns, Mig 38): ProposalFeatures.cs:403-486 ApproveHandler = flatten Steps.OrderBy(Order).SelectMany(Levels.OrderBy(Order)) global level index → allLevels[CurrentApprovalLevelOrder-1] → actor match Level.ApproverUserId==uid||Admin → UPSERT LevelOpinion → advance++/DaDuyet. ⚠️ ApprovalWorkflow.cs:72 nói Level KHÔNG OR-of-N (1 ApproverUserId/Level) — KHÁC memory cũ "OR-of-N", verify lại. Gap: 4 bảng {Module}LevelOpinion mới (Mig 41+), 3 route/controller, 4 seed WF, FE Detail+Opinion (chỉ có WorkflowAppsListPage chung). ⚠️ enum ApplicableType THIẾU Travel (có Leave=5/OT=6/Vehicle=7/ItTicket=8); ExtendApplicableTypeForWorkflowApps mig empty Up/Down (enum-only). Tag [pre-flight, p11-a, workflowapps].