Skip to main content

Post-close corroboration addendum (2026-07-19)

Extracted verbatim from the sprint plan per the terse-authoring rule (.claude/repo-local/delivery-terse-authoring.md): full evidence lives here; the sprint doc keeps a one-line-per-finding summary.

Retro action #2 (above) was actioned during Sprint 2 planning: qontak-billing and report-worker were both cloned as sibling checkouts and manually walked (branch search + git merge-base --is-ancestor against master) to close the evidence gap these findings were held on. This addendum updates F-13/F-14/F-20/F-21's status as of 2026-07-19 — the pass-07 figures above are left untouched as the historical record at close; this is a dated update layered on top, per the amend-mode date-stamping convention.

  • F-13 (self-subs BIF-8803/8804) — fully resolved 2026-07-19, both are genuine redirects:
    • BIF-8803 (Task 2.3, SelfSubscriptions initiate-proxy) — Jira comment (2026-07-13, Grehasta Mahardika): "already handled using this ticket: BIF-8792 ... Task 2.3: SelfSubscriptions Grape resource + routes (SS-S02, SS-S03, SS-S04) Done." Verified: hub-service commit 13b414d5a/merge a2e237fc8 and hub-core commit 163a271db2 are a verbatim match to Task 2.3, confirmed merged to origin/production in both repos since 2026-07-10, approved by Hafriz Resa Damarsidi.
    • BIF-8804 (Task 2.4, modpanel disable_self_renewal gate) — fetched directly from Jira: comment (2026-07-13, Grehasta Mahardika): "will move to wont fix because already done in thic ticket: BIF-8793." BIF-8793 ("[BE hub-service] Task 2.4: modpanel PUT disable_self_renewal + enforce gate") is Done on the board. Verified across all three repos the task-breakdown names: hub-service (ada19b041, param persistence — merged origin/production), hub-core (d3afe1ce8f, the actual 422 enforcement gate + UpdateRemainingBalance threading — merged origin/production), and moderator-be (e454ff13/24fe5eef, permit+forward — merged origin/master). Full task genuinely shipped.
    • Both Won't Fix resolutions were correct — the work shipped under different tickets than the breakdown's canonical ids. F-13 is closed, not just downgraded.
    • Task 2.2 clarification (previously flagged as possible duplication): BIF-8791 ("hub_core") and BIF-8802 ("hub-service") are not a duplicate/redirect pair — Task 2.2 is a genuine two-repo split per the breakdown. BIF-8791 is Done and confirmed merged to hub-core production. BIF-8802 is In Deployment on the board, and code-verified as accurate, not stale — its merge commit (9b8350077) is on origin/release/0.267.0 but not yet an ancestor of origin/production. No further engineering work remains on Task 2.2, only the 0.267.0 release landing.
  • F-14 (self-topup BIF-8708/8709) — still open, reconfirmed, now stronger: qontak-billing checked (was the missing repo — "PARTIAL" above). Zero trace across all three repos in scope (hub-chat, hub-core, qontak-billing). Treat the board Done status as very likely stale unless Madani/TPM confirms a true no-diff fix.
  • F-20 (Meta WhatsApp Pricing) — substantially resolved: Task 1.2 (BIF-8808, qontak-billing, PRs #293/#294) and Task 1.3 (BIF-8809, report-worker, PR #165, approved by Grehasta Mahardika) are both confirmed merged to master. Phase-1 is genuinely 3/5 issues corroborated (~6.0/9.0 Phase-1 md), not the 1/5 read at close. Only Task 1.4 (descoped/Won't Fix) and Task 1.5 (FE, gated on the open pricing decision) remain unconfirmed — both are legitimate gates, not code-evidence gaps. Progress can move from the held 33% (4.0/12) once Addo/TPM confirms.
  • F-21 (Downgrade Webhook) — split verdict, now a confirmed conflict on half: Phase 1 (schema — triggers_downgrade, downgrade_events/downgrade_schedules) is confirmed merged (PR #296, BIF-8813). Phase 2 (logic & workers — DowngradeNotificationWorker, the compareComponents Day-0 flow, BIF-8815) is a confirmed unmerged branch (merge-base --is-ancestor = false against master) — if any of the 3 in-sprint issues covering that logic reads Done on the board, that is now a genuine board/code conflict, not an evidence gap, and needs Madani/TPM to reconcile directly rather than another recheck pass.
  • New, unresolved: the same BIF-8813/PR #296 merge also carries wa_balance_holds / wa_reconciliation_batches migrations (authored by Madani) that map to neither Downgrade Webhook's nor Meta WhatsApp Pricing's task-breakdown. Flagged to TPM as unexplained scope — not attributed to either initiative pending an answer.

These updates are also reflected forward in 2026-07-21-sprint-02.md's queues and Deviations section, which is where the live, actionable versions of this evidence now live.