Request:
Run the scheduled HK IPO analyst refresh as of 2026-06-24T07:00:26Z, refresh online archive facts first, rebuild the analysis dataset, write the latest Chinese broad candidate report, mirror it to reports/README.md, and preserve stage discipline.
Changes:
- Refreshed HKEX English and Chinese current-listing pages, rolling recent-listing coverage, current HKEX document searches, VBKR/Jieli live T0.95 market heat, ipohk external history, Yahoo same-day D1 evidence, and A/H quote/FX evidence.
- Archived 5 still-actionable T0_95_final_heat rows for 00668, 02697, 03952, 06715, and 06915 while leaving post-deadline names on their last pre-deadline heat snapshots.
- Rebuilt analysis_model_v0_dataset.csv and related snapshots for 2026-06-24T07:00:26Z.
- Updated reports/2026-06-24_latest_ipo_candidates_analysis.md in Simplified Chinese and mirrored identical content to reports/README.md.
- Kept unofficial heat in ipo_market_heat only, preserved official T1 demand from HKEX allotment-result sources, and kept 02335/06106 D1 rows labelled as in-session rather than final D1 confirmation.
Verification:
- git diff --check
- git diff --cached --check
- Rebuilt analysis dataset for 2026-06-24T07:00:26Z with 312 rows
- Python check that reports/README.md matches the dated report and all unresolved current/recent tickers are covered in the risk table
- Python check that 5 current-run heat rows are T0_95_final_heat with provider VBKR/Jieli and active heat tickers have no official ipo_demand rows
- Python check that 02335 and 06106 official T1 fields match HKEX allotment results
- Python check that 89 source_refs archived at 2026-06-24T07:00:26Z use repo-relative paths, files exist, and hashes match
Next useful context:
- At 15:00 HKT, only 00668, 02697, 03952, 06715, and 06915 remained still-actionable by public subscription timetable; 06715 crossed 10x heat and upgraded to a small secondary candidate.
- 02335 and 06106 have refreshed 2026-06-24 Yahoo D1 rows before the Hong Kong close; treat them as in-session execution evidence, not final D1 review labels.
- T2 grey-market remains a source-strategy data_gap for recent June IPOs without approved reproducible evidence.
Request:
Run the scheduled HK IPO analyst refresh as of 2026-06-24T05:23:44Z, refresh online archive facts first, rebuild the analysis dataset, write the latest Chinese broad candidate report, mirror it to reports/README.md, and preserve stage discipline.
Changes:
- Refreshed HKEX English and Chinese current-listing pages, rolling recent-listing coverage, the due HKEX document task for 02667, VBKR/Jieli live market heat, ipohk external history, Yahoo price-performance evidence, and A/H quote/FX evidence.
- Archived 5 still-actionable T0_95_final_heat rows for 00668, 02697, 03952, 06715, and 06915 while leaving post-deadline names on their last pre-deadline heat snapshots.
- Rebuilt analysis_model_v0_dataset.csv and related snapshots after the archive refresh.
- Wrote reports/2026-06-24_latest_ipo_candidates_analysis.md in Simplified Chinese and mirrored the same content to reports/README.md.
- Kept unofficial heat in ipo_market_heat only, preserved official T1 demand from HKEX allotment-result sources, and labeled 02335/06106 D1 rows as in-session rather than final D1 confirmation in the report.
Verification:
- git diff --check
- Rebuilt analysis dataset for 2026-06-24T05:23:44Z
- Python check that reports/README.md matches the dated report and all unresolved current/recent tickers are covered
- Python check that 5 current-run heat rows are T0_95_final_heat with provider VBKR/Jieli and active heat tickers have no official ipo_demand rows
- Python check that 02335 and 06106 official T1 fields match HKEX allotment results
- Python check that 68 source_refs archived at 2026-06-24T05:23:44Z use repo-relative paths, files exist, and hashes match
- Python check that analysis_model_v0_dataset.csv has 312 rows and the expected analysis_as_of
Next useful context:
- At 13:23 HKT, the 6/24 10:00 deadline group was already closed, leaving only 00668, 02697, 03952, 06715, and 06915 as still-actionable subscription names.
- Yahoo returned 02335 and 06106 D1 daily rows before the Hong Kong close; treat them as D1 in-session execution evidence, not final D1 review labels.
- T2 grey-market remains a source-strategy data_gap for recent June IPOs without approved reproducible evidence.
Request:
Run the scheduled HK IPO analyst refresh as of 2026-06-23T23:00:03Z, refresh online archive facts first, rebuild the analysis dataset, write the latest Chinese broad candidate report, mirror it to reports/README.md, and preserve stage discipline.
Changes:
- Refreshed HKEX current-listing pages, current HKEX document searches, VBKR/Jieli T0.95 market heat, ipohk external history, and A/H quote evidence at the new as-of timestamp.
- Preserved unofficial subscription heat only in ipo_market_heat and kept official T1 demand fields sourced from HKEX allotment-result documents.
- Marked due or Hong Kong-time elapsed T2 grey-market source-strategy gaps for 01392, 02335, 02667, 06067, 06106, 06132, and 06675.
- Backfilled 06675 stock_short_name from archived Yahoo shortName evidence and rebuilt analysis_model_v0_dataset.csv.
- Updated reports/2026-06-23_latest_ipo_candidates_analysis.md and mirrored the same content to reports/README.md with actionable ranking, fundamentals, unresolved-D1 risk/reward coverage, closed/waiting names, A/H overlay, recent 30-day review, guardrails, and sources.
Verification:
- git diff --check
- Rebuilt analysis dataset for 2026-06-23T23:00:03Z
- Python check that reports/README.md matches the dated report and required current/recent tickers are present
- Python check that 23:00Z heat has 8 T0_95_final_heat rows and active heat tickers have no official ipo_demand rows
- Python check that 02335 and 06106 official T1 fields match HKEX allotment results
- Python check that 77 source refs archived at 2026-06-23T23:00:03Z use repo-relative paths, files exist, and hashes match
- Python check that T2 data gaps are marked for 01392, 02335, 02667, 06067, 06106, 06132, and 06675
Next useful context:
- VBKR/Jieli heat values for the 8 still-actionable names were unchanged from 15:00Z but now have a fresh 23:00Z observed_at.
- 02335 and 06106 have official T1 demand, but D1 had not opened by the report as-of time; T2 remains a source-strategy data_gap.
- 00901 Yahoo D1 fetch still returns 404; ipohk remains only a third-party cross-check.
Request:
Run the scheduled HK IPO analyst refresh as of 2026-06-23T15:00:19Z, refresh online archive facts first, rebuild the analysis dataset, write the latest Chinese broad candidate report, mirror it to reports/README.md, and preserve stage discipline.
Changes:
- Refreshed HKEX current-listing pages, VBKR/Jieli T0.95 market heat, ipohk external history, A/H quote evidence, and current HKEX document searches.
- Archived official HKEX allotment-result PDFs and extracted text for 02335 and 06106; parsed official T1 demand into ipo_demand without copying market heat into official fields.
- Rebuilt analysis_model_v0_dataset.csv and refreshed sync/source snapshots.
- Updated reports/2026-06-23_latest_ipo_candidates_analysis.md and mirrored the same content to reports/README.md, including current ranking, fundamentals, unresolved-D1 risk/reward table, closed/waiting names, 30-day review, guardrails, and sources.
Verification:
- git diff --check
- Rebuilt analysis dataset for 2026-06-23T15:00:19Z
- Python check that reports/README.md matches the dated report and required new facts are present
- Python check that 15:00Z heat has 8 ipo_market_heat rows and current actionable names have no official ipo_demand rows
- Python check that 02335 and 06106 official T1 fields match HKEX allotment results
- Python check that 77 source refs archived at 2026-06-23T15:00:19Z use repo-relative paths, files exist, and hashes match
Next useful context:
- 02335 and 06106 now have official T1 demand, but D1/T2 remain data_gap until listing-day evidence is archived.
- 00901 Yahoo D1 fetch still returns 404; ipohk remains only a third-party cross-check.
Request:
Update the latest HK IPO candidate report with fresh market heat and enforce the new Chinese-name report format.
Changes:
- Archived the 2026-06-23 10:59Z VBKR/Jieli heat snapshot and HKEX Chinese current listings page.
- Backfilled current IPO Chinese names from the HKEX Chinese page, leaving 06228 as data_gap because the official Chinese page lists only the foreign issuer name.
- Refreshed A-share and HKD/CNY raw quote JSON for the A/H overlay.
- Rebuilt analysis snapshots and refreshed the 2026-06-23 latest report plus reports/README.md with dedicated Chinese-name columns.
Verification:
- git diff --check
- Confirmed reports/README.md and reports/2026-06-23_latest_ipo_candidates_analysis.md are identical.
- Checked the report contains the 2026-06-23T10:59:36Z timestamp and latest heat source path.
- Checked old English company-name labels are absent from the refreshed reports.
Next useful context:
- 06228 remains a Chinese-name data_gap unless an official Chinese issuer name source appears.
Request:
- Regenerate the latest HK IPO candidate report.
- Include current IPO candidate refresh, updated subscription heat, A/H dual-listed pricing analysis for 03661 and 01688, and T2 grey-market review context.
Changes:
- Refreshed HKEX current-listing, VBKR/Jieli T0.95 heat, ipohk external history, sync-state, and analysis-model snapshots as of 2026-06-23T08:53:26Z.
- Archived raw Yahoo chart evidence for 300661.SZ, 002600.SZ, and HKD/CNY so the A/H discount overlay has a reproducible local source.
- Regenerated reports/2026-06-23_latest_ipo_candidates_analysis.md and mirrored it to reports/README.md.
- Added the generated model v0 report from the rebuilt analysis dataset.
- Marked T2 grey-market evidence quality explicitly: ipohk grey-market returns are Tier 3 historical summaries, while newer June IPOs remain data_gap.
Verification:
- Confirmed reports/README.md matches the dated latest report with cmp.
- Ran git diff --check and git diff --cached --check.
- Verified all repo-relative paths referenced by the latest report exist.
- Verified source_refs paths are repo-relative, existing, and hash-matching.
- Recomputed A/H overlay values from archived raw JSON: 03661 discount about 46.2%, 01688 discount about 45.1%.
Next useful context:
- The 2026-06-23T08:53:26Z heat snapshot leaves the same 8 currently actionable candidates as the 07:00 refresh, with small heat updates.
- 02335 and 06106 still have no official T1 demand rows in the project archive.
- A/H discounts are valuation anchors, not direct arbitrage, because the A and H shares are not fungible.
Request:
- Run the scheduled hk-ipo-analyst refresh as of 2026-06-23T07:00:25Z.
- Refresh the latest IPO candidate universe and online facts through the archivist before analysis.
- Rebuild the analysis dataset and publish the latest broad candidate report in Simplified Chinese.
Changes:
- Archived the HKEX current new-listing page for 2026-06-23 and added the new 00668 prospectus plus extracted text.
- Archived a fresh VBKR/Jieli T0.95 market-heat snapshot with 8 still-actionable rows while leaving same-day closed names out of the live T0.95 set.
- Archived 06067 and 06132 D1 price-performance responses, refreshed ipohk external history, and archived the 06067 2026-06-23 allotment-results clarification PDF.
- Rebuilt data/snapshots/analysis_model_v0_dataset.csv after the archive refresh.
- Wrote reports/2026-06-23_latest_ipo_candidates_analysis.md and mirrored the same content to reports/README.md.
Verification:
- Ran archive_hkex_current_new_listings.py, archive_hkex_documents.py, backfill_t1_demand_from_text.py, archive_t0_5_market_heat.py, archive_price_performance.py, archive_ipohk_history.py, extract_pdf_text.py, update_sync_state.py, and build_analysis_dataset.py with as-of 2026-06-23T07:00:25Z.
- Confirmed reports/README.md matches the dated report with cmp.
- Ran git diff --check and git diff --cached --check.
- Checked source_refs paths are repo-relative, existing, and hash-matching.
- Checked the latest 8 T0.95 live heat rows remain separate from official T1 demand rows.
Next useful context:
- As of 2026-06-23T07:00:25Z, 02335 and 06106 still have no archived official T1 demand rows and remain T1 data_gap names.
- 06067 D1 was positive at about +30.0%, while 06132 D1 was negative at about -49.6%, reinforcing the 18A-B risk guardrail.
- The 2026-06-23 HKEX current page shows 00668 as a new live candidate; its first VBKR/Jieli heat snapshot was only 0.32x.
Request:
- Run the scheduled hk-ipo-analyst refresh as of 2026-06-22T23:00:18Z.
- Refresh the IPO candidate universe and latest online facts through the archivist before analysis.
- Rebuild the analysis dataset and publish the latest broad candidate report in Simplified Chinese.
Changes:
- Re-archived the HKEX current new-listing page at the requested as-of time and refreshed sync state.
- Archived a fresh VBKR/Jieli T0.95 market-heat snapshot for the 13 actionable candidates while keeping unofficial heat in ipo_market_heat.
- Refreshed ipohk external history, reran official HKEX document/T1 demand and price-performance refreshes, and rebuilt analysis_model_v0_dataset.csv.
- Rewrote reports/2026-06-22_latest_ipo_candidates_analysis.md for the 23:00Z snapshot and mirrored it to reports/README.md.
Verification:
- Ran archive_hkex_current_new_listings.py, archive_hkex_documents.py, backfill_t1_demand_from_text.py, archive_t0_5_market_heat.py, archive_price_performance.py, archive_ipohk_history.py, and build_analysis_dataset.py with as-of 2026-06-22T23:00:18Z.
- Confirmed reports/README.md matches the dated report with cmp.
- Ran git diff --check and git diff --cached --check.
- Checked source_refs paths are repo-relative, existing, and hash-matching.
- Checked the latest 13 T0.95 live heat rows remain separate from official T1 demand rows.
Next useful context:
- The 23:00Z VBKR/Jieli heat values matched the prior 15:47Z values for the 13 active candidates.
- No new official T1 rows were available for 06106 or 02335 as of 2026-06-22T23:00:18Z; keep them as T1 data_gap names.
- Yahoo price refresh wrote no new internal D1 rows; 00901 remains an internal D1 data_gap.
Request:
- Run the scheduled hk-ipo-analyst refresh as of 2026-06-22T15:47:32Z.
- Refresh the IPO candidate universe and network facts through the archivist before analysis.
- Rebuild the analysis dataset and publish the latest broad candidate report in Simplified Chinese.
Changes:
- Archived the HKEX current new-listing page, new official allotment-result PDFs for 06067 and 06132, extracted text, a fresh VBKR/Jieli T0.95 market-heat snapshot, ipohk external history, and recent Yahoo price-performance responses.
- Updated structured SQLite facts and CSV snapshots, including official T1 demand for 06067 and 06132 while keeping live subscription heat in ipo_market_heat.
- Rebuilt data/snapshots/analysis_model_v0_dataset.csv after the archive refresh.
- Rewrote reports/2026-06-22_latest_ipo_candidates_analysis.md and mirrored the same content to reports/README.md.
Verification:
- Ran archive_hkex_current_new_listings.py, archive_hkex_documents.py, backfill_t1_demand_from_text.py, archive_t0_5_market_heat.py, archive_price_performance.py, archive_ipohk_history.py, and build_analysis_dataset.py with as-of 2026-06-22T15:47:32Z.
- Confirmed reports/README.md matches the dated report with cmp.
- Ran git diff --check and git diff --cached --check.
- Checked source_refs paths are repo-relative and existing.
- Checked the latest 13 T0.95 live heat rows remain separate from official T1 demand rows.
Next useful context:
- 06067 and 06132 now have official T1 demand in the archive; 06106 and 02335 remain T1 data_gap names as of this run.
- The 15:47Z VBKR/Jieli live heat values matched the earlier 13:57Z values for active candidates.
- Price refresh still has provider gaps for some historical tickers, including internal D1 price data for 00901.
Request:
- Update the current HK IPO candidates after subscription multiples changed again.
- Refresh the candidate ranking and defensive risk/reward view.
Changes:
- Archived the 2026-06-22T13:57:38Z VBKR/Jieli T0.95 market-heat snapshot.
- Re-archived the HKEX current new-listing page and rebuilt snapshots/model dataset.
- Updated the June 22 candidates report with the latest heat multiples, ranking, execution guidance, and defensive score table.
- Refreshed the model report analysis timestamp.
Verification:
- Ran git diff --check and git diff --cached --check.
- Confirmed SQLite PRAGMA integrity_check = ok.
- Parsed changed CSV snapshots and confirmed 13 latest T0_95_final_heat rows.
- Checked the current candidate ranking against data/snapshots/analysis_model_v0_dataset.csv.
Request:
- Update the latest Hong Kong IPO candidate list and rescore it based on subscription multiples.
Changes:
- Archived the 2026-06-22 HKEX Main Board New Listing Information page, adding 02697, 03952, 06715, and 06915 to the current candidate set.
- Archived and extracted the four new prospectuses, refreshed current HKEX document facts, and rebuilt the v0 analysis dataset to 311 rows.
- Archived a 2026-06-22T08:55:00Z VBKR/Jieli market-heat snapshot and wrote only still-actionable T0.95 rows to avoid look-ahead leakage for already-closed IPOs.
- Improved prospectus date parsing for split weekday/month text, glued noon/commence phrases, and current new-listing expected listing-date updates.
- Added a Chinese 2026-06-22 latest IPO report ranking candidates after the subscription-multiple overlay.
Verification:
- Ran py_compile for archive_hkex_documents.py, archive_t0_5_market_heat.py, archive_hkex_current_new_listings.py, and build_analysis_dataset.py.
- Re-ran HKEX current-page seeding, document archiving, market-heat archiving, and analysis dataset build as of 2026-06-22T08:55:00Z.
- Ran git diff --check and git diff --cached --check.
- Ran SQLite integrity_check and foreign_key_check.
- Verified source_refs paths, file existence, and SHA-256 hashes.
Next useful context:
- 01956 is the only current candidate with both strong T0 structure and >100x actionable heat in this snapshot.
- Recheck 03952 and 06715 near the 2026-06-25 cutoff; their structure is strong but 2026-06-22 heat is below 10x.
- Official T1 allotment facts for 06067 and 06132 were still unavailable at this archive timestamp.
Request:
- Use the project analyst workflow to analyze the latest upcoming Hong Kong IPO candidates.
Changes:
- Refreshed recent HK IPO target coverage through 2026-06-17 and archived current HKEX source updates.
- Archived 06675 allotment results and D1 Yahoo price performance for boundary-case review.
- Archived a 2026-06-17 T0.5 VBKR/Jieli market-heat snapshot for still-actionable 02335 and 06106.
- Rebuilt the v0 analysis dataset and snapshots at 2026-06-17T08:20:00Z.
- Added a Chinese horizontal analyst report ranking 06106, 02335, 06132, 06067, 01392, with 06675 separated as a T1/D1 review sample.
Verification:
- Ran SQLite PRAGMA integrity_check and foreign_key_check.
- Ran git diff --check and git diff --cached --check.
- Confirmed report source paths exist.
Next useful context:
- 06106 is the top still-actionable T0.5 candidate at this as-of time.
- 02335 needs another pre-deadline heat sample before a stronger call.
- 01392, 06067, and 06132 are now mainly waiting for T1 official allotment results.
Request:
- Test whether subscription-period T0.5 market heat data can be captured and incorporated into IPO analysis.
Changes:
- Add an ipo_market_heat table for non-official market-heat snapshots.
- Add a VBKR/Jieli archive script for expected margin subscription multiples.
- Archive the 2026-06-15T18:40:00Z heat snapshot for 01392, 02335, 06067, 06106, and 06132.
- Add an experimental T0.5 overlay rule file and a Chinese cross-IPO trial report.
- Update archivist and analyst skills so T0.5 remains separate from official T1 allotment demand.
Verification:
- .venv/bin/python -m py_compile scripts/archive_t0_5_market_heat.py scripts/build_analysis_dataset.py scripts/update_sync_state.py
- Python sqlite3 PRAGMA integrity_check returned ok and foreign_key_check returned zero rows.
- Confirmed 5 ipo_market_heat rows and 5 t0_5_market_heat source_refs for the frozen snapshot.
- git diff --cached --check
Next useful context:
- T0.5 data is non-official and should be resampled during the subscription window, then compared against T1 official allotment results.
geometrybase