CDP Sprint 1 (2026-Q3) — Jira reconciliation, pass 8 (2026-07-16)

Jira reconciliation — recheck (2026-07-16, pass 8)

Eighth pass, run via sprint:reconcile (qontak-docs, npm run sprint:reconcile -- --team cdp --json), same day as pass 7 — triggered by a targeted ask to find unlinked/unparented Jira items in sprint 21382. Sanity check first: sprint = 21382 AND parent is EMPTY → 0 results — no issue in the sprint is missing a parent outright. The engine's own epic-wiring-drift row is the only structural gap, and this pass verifies it directly against live Jira.

Status of prior findings: no re-verification of F-1 through F-15 this pass (unchanged since pass 7, same day); this pass is scoped to the wiring question only.

New finding:

16. F-16 epic-wiring-drift — NEW (pass 8): TF-3524 ("[QA] Create Automation Test — Loyalty cases", assignee Muhammad Izzul, status In Progress) parents to Epic TF-3521 ("26Q3 CDP - QA DORA" in Jira; referred to in the 2026-07-16 Sprint 2 doc / ce31079 commit as the CDP "QA BAU" epic). TF-3521 is not registered on any CDP initiative README's jira_epic, the roadmap, or Sprint 1's ## Initiative index — the sole off-map parent among sprint 21382's 73 issues (every other issue resolves to a tracked Epic: TF-3184/3185/3181/2544/3183/3477/3182/ 3500/3359). Confirmed via direct Jira lookup: TF-3521 was created 2026-07-16 15:52 by the repo's own TPM/PM (Hilmi Muktitama), ~2 hours before the commit that documents it — i.e. genuinely new, not a stale mapping.

    This also resolves the read on F-9. F-9 has stood since pass 3 as "informational, accepted: Izzul reports 0 sprint issues every pass — expected, deliberately-benched backup QA." That premise no longer holds: Izzul has real, in-progress work (TF-3524); it was simply invisible to the reconcile engine's initiative-scoped counting because TF-3521 isn't tracked. Izzul was never idle this pass — the epic was unregistered. F-9 is superseded by F-16 below.

    Owner: TPM. Fix applied this pass: recorded as an off-plan Deviation (same convention as the Engineering Initiatives bucket TF-3500/A4 pull-in) rather than a full initiative-folder registration, since it's a single BAU item so far — revisit and register a dedicated bucket initiative + roadmap row if TF-3521 accrues more sprint work. Sprint 1's Izzul queue section was updated to show TF-3524 instead of "no active queue."

    Not in scope this pass (deferred, not forgotten): TF-3521's Jira name ("26Q3 CDP - QA DORA") doesn't match the descriptive label the Sprint 2 doc invented for it ("QA BAU — Automation Coverage & Deployment Support") — verified against sibling squads' equivalent epics (Bifrost BIF-8869 = "26Q3 QA DORA Metrics", Chatbot BOT-4609 = "26Q3 QA Automation & Delivery"): TF-3521's actual name is consistent with the cross-squad convention, so this is a docs-text accuracy issue, not a Jira problem. Left open for a future pass/decision — no rename or doc-text edit made here.

What this pass fixes: Sprint 1's ## Initiative index/Deviations now account for TF-3521/TF-3524; Izzul's queue section reflects real board state; the ledger's F-9 is marked superseded and F-16 records the underlying wiring gap with its resolution.

Downstream propagation worklist (pass 8)

#	Finding	Canonical artifact to reconcile	Owning tool / skill
16	TF-3524 off-map under TF-3521 (epic-wiring-drift)	cdp/delivery/sprints/2026-07-07-sprint-01.md ## Deviations + Izzul queue	hand-edit, done this pass
16	TF-3521 name mismatch (docs vs Jira) — deferred	Sprint 2 doc + roadmap prose, or Jira Epic rename	TPM decision, not yet actioned

Always run npm run lint:delivery (DOCS_ROOT pointed at this repo) after any propagation edit.