diff --git a/.project-docs/30-worklog/tasks/20261008-production-review-9e7b.md b/.project-docs/30-worklog/tasks/20261008-production-review-9e7b.md
index cb8256e..48a0826 100644
--- a/.project-docs/30-worklog/tasks/20261008-production-review-9e7b.md
+++ b/.project-docs/30-worklog/tasks/20261008-production-review-9e7b.md
@@ -231,3 +231,12 @@ Read: memory-index, project-positioning, current-state latest September sections
 - User challenged the long wait. Same-task ownership/context were resumed and verified again. At13:35:22UTC, details reached195/209 (29 additional details/rates/profiles since13:31:48); latest response ended13:35:21. Successful request mean times: detail2.92s, rate2.20s, profile2.21s. This supports serial query latency as the dominant observed wait, without blaming Oracle for all overhead or claiming the complete request was stalled. At13:36:39, details reached205/209.
 - Actual completion: last successful source response13:37:25UTC;209 details,209 rates,209 profiles,6 search pages including final recheck,1 room-calendar request;634 HTTP200 plus1 transport failure/retry. Capture result collection_complete=true,209 records,error=null. At13:37:31UTC (21:37 local computer time), the original request transitioned to needs_review after about26m13s. Authenticated read-only API reports existing price review open, revision0,0/5 completed. Missing raw observations do not imply5 source-field prompts: eligible source review passed into the existing price review.
 - Fresh temporary authenticated browser verified selected9/16, “等待人工价格复核”, enabled “下载并处理” and “查看价格复核”, and0/5 review progress. No price form was edited or finalized. Closed the verification tab; existing user tabs and requests were untouched. This establishes both natural task completion and download-button recovery, with no restart/refetch/code change/test run. Speed optimization and visible progress are potential follow-ups; the current26-minute duration is observed poor product performance, not evidence of a deadlock.
+
+## Same-task Follow-up: September16 XML Price Comparison
+
+- User supplied `/Users/chillishark/Downloads/res_detail_0916.XML` and asks whether the5 manual-price review items were already0 in the original XML. Concurrent gate Passed for the same feature task/codex/owned checkout/base2417b1a, no peers. Retained required Project Context Loaded and current task record apply; Planning gate Passed for read-only comparison. No broad population-policy alignment, report generation, price substitution or new Oracle call is inferred. Read the report skill for relevance, but its generation workflow is not invoked: this request is diagnostic XML/source comparison, not a new daily/monthly output.
+- Original XML880858 bytes, SHA256 `bc99252b1cb51f9929dfe66f62cb4c4bd572484ef4827fc0e59261fd934c26df`,184 reservations, all arrival16-09-26/statusCKOT/hotel57106. Original bytes remain unchanged. There are2 XML0-price records with COMP/HOUSE1, neither belongs to the45 reservations behind these5 price-review keys.
+- Exact confirmation-number join against the frozen9/16 structured-result price_unmatched records: LIAN TAI/GRP1/API0 affects2, XML matches0; LIAN TAI/WHO1/API0 affects6, matches0; LIAN TAI/WHO2/API0 affects1, matches0; LIAN TAI/WHO2/API1000 affects35, all35 match XML WHO2/1000/CKOT; QBD/GRPA1/API0 affects1, matches0. Thus5 review groups cover45 reservations, not5 reservations. The10 zero-priced source reservations are all CheckedOut in the saved API source but absent from this XML. The file cannot establish their XML price or why Oracle's exported population omitted them; no cancellation/status-export assumption is justified.
+- Independent audit normalized XML company identities and checked the frozen fixed-price loader. LIAN TAI+WHO2+1000 has no exact reference entry (available source amounts1100,1400,1600,1800,2400); all4 API0 keys also have no exact fixed entry. Existing shared rule matches company+rate+Opera amount and requests a nightly processed price for unmatched keys. It is not a generic “0 price is invalid” rule, and source0 does not imply final processed price0. XML's WHO2/1000 values agree with the API, while the fixed-price mapping is missing for that combination.
+- Authenticated read-only review at start: open revision1,1/5 completed, user had already saved0 for LIAN TAI/GRP1. Kept this and every other actual decision untouched. No source/rule/reference file, service state, Finance/monthly data or actual report was modified; no Oracle read, test run or full report processing occurred. Private per-reservation diagnostic JSON: `production-validation-20261007/sept16-price-xml-comparison-20261008.json`. Guest names and raw responses remain outside Git.
+- Follow-up/promotion: explain grouped price issues separately from individual reservations, and distinguish a missing fixed processing-price mapping from0 source values and absent XML membership. Broad source-population alignment remains deferred; do not automatically zero, exclude or import the10 absent reservations based on this comparison. Any change to the permanent reference price requires an explicit business value; staff may still save a per-report0 override where they have confirmed it.