vietreport-admin/solution-erp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[CLAUDE] Docs: S148 — gỡ note dạng chuỗi khỏi class_repeat.counts (map chỉ chứa SỐ)
All checks were successful
Deploy SOLUTION_ERP / build-deploy (push) Successful in 5m28s
Details

Browse Source 
Lead nhét _backfill_S148 (string) vào counts{} nơi chỉ được có int → consumer đọc
map số sẽ vấp (chính filter -ge 3 của lead vừa in nó ra như một 'class chạm JUMP').
Chuyển note lên cấp cha class_repeat. counts{} nay 8 key, toàn int.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
pqhuy1987
2026-07-24 18:23:13 +07:00
parent aaf95efba7
commit 2f39a7e22c
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
.claude/governance/.session-counter.json
View File

@ -1,4 +1,4 @@
{
{
"_note": "Harness-24 (S121, 2026-07-15, wave adap-6-broadcast W2) session counter for lead-self-audit cadence. THIS FILE HOLDS STATE ONLY - every cadence NUMBER lives in .claude/agent-memory/memory-budget.json -> h24_cadence (single-source, H24 section 2(3)). Never copy the digits here.",
"_why_content_based_not_blind_increment": "Owner-decisions-15-07-2026 Q2, trap found by lane E: the Stop-hook .claude/hooks/wal-flush.ps1 auto-commits on EVERY turn-boundary. A blind counter++ would therefore tick many times per real session and the cadence would be meaningless. So a tick is CONTENT-ADDRESSED: it is keyed on the repo HEAD sha at tick time. Ticking twice on the same HEAD is a NO-OP, which is what makes running /session-start twice in a row safe (acceptance: 2x /session-start on the same HEAD => +1, not +2).",
"_why_this_path": "Deliberately .claude/governance/ and NOT .claude/agent-memory/. The hook covers exactly four paths (.claude/WAL.md, agent-memory/, workflows/runs/, sessions/ - the 4th added S146 together with the session-model logic-session folder); .claude/governance/ is OUTSIDE all four, so this file is never swept into a 'wal:' commit as a side effect. Compare .claude/agent-memory/.mfe-state.json, which IS inside the hook's reach - a counter must not live there. RE-VERIFIED at S146 when the 4th path landed: the claim is about the DIRECTORY, and .claude/sessions/ != .claude/governance/, so widening the hook did NOT invalidate it - only the count changed. Any future widening of wal-flush.ps1 MUST re-check this sentence, not just bump the number.",
@ -28,11 +28,11 @@
"gap-carry-dropped": 3,
"gap-decision-sunk": 3,
"view-stale-status": 0,
"_backfill_S148": "S146 audit CHAY THAT (force-fire P=8) nhung KHONG co entry trong history[] va class_repeat KHONG duoc cap-nhat — lead-omission-auditor bat @S148 (FLAG-3 gap-decision-sunk), lead verify doc-lap 2 nguon: (a) history[] nhay S145->S147, (b) sub-lead-omission-S146.md ghi nguyen-van 'tally 2 -> 3 CHAM JUMP (S140 x1 -> S143 x2 -> S146 x3)'. Tally duoi day DA VA cho khop chuoi that S140->S143->S146->S148. HE-QUA da xay ra: ghi-chu tick S148 ket-luan 'cao nhat = 2 < 3 => khong jump' — DOC TU SO TIEN-S146, sai. Jump da fire tu S146 ma khong ai biet; no duoc tra TINH-CO boi force-fire P=1 @S148 (ly-do khac).",
"gap-owner-specifics": 4,
"gap-carry-aged": 1,
"gap-incident-unrecorded": 0
}
},
"_backfill_S148": "S146 audit CHAY THAT (force-fire P=8) nhung KHONG co entry trong history[] va class_repeat KHONG duoc cap-nhat — lead-omission-auditor bat @S148 (FLAG-3 gap-decision-sunk), lead verify doc-lap 2 nguon: (a) history[] nhay S145->S147, (b) sub-lead-omission-S146.md ghi nguyen-van 'tally 2 -> 3 CHAM JUMP (S140 x1 -> S143 x2 -> S146 x3)'. Tally duoi day DA VA cho khop chuoi that S140->S143->S146->S148. HE-QUA da xay ra: ghi-chu tick S148 ket-luan 'cao nhat = 2 < 3 => khong jump' — DOC TU SO TIEN-S146, sai. Jump da fire tu S146 ma khong ai biet; no duoc tra TINH-CO boi force-fire P=1 @S148 (ly-do khac)."
},
"history": [
{