The two upstream files stay verbatim in scripts/. The path helper moved into a subdirectory so the root file budget rises by only those two. Niryaana Shoola is a placeholder until its rule is corrected. Targeted tests passed.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Move the three local glue modules into scripts/marriage_match_service and point the invocation matrix at marriage_match.py. The upstream engine file stays where it is.
Checked the file-count and oracle contracts, the marriage-match tests, and a Windows full pytest. The 176 failures are python3-missing and path-separator noise; the named contracts passed.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
The western comparison layer was invoked without its files, so it stayed blocked. Navamsha and Niryaana Shoola are filled from the native rashi dashas.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Sync8 changed ashtakavarga.py, dasha_analyzer.py and narayana_dasha.py,
which are in the frozen scoring identity. Sealed (20) and reported-offset
(900) replays match the scoring_perf_2026_10_13 records trial for trial,
so ERR-110 stores a new record. Prior records are kept byte-for-byte.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HwMNZEzdzmrbPhuRNmZe3C
Polar day or night left determine_daytime without a sunrise, so the whole
avasthas module failed after the PVR activity formula was wired in. Only
the Shayanadi cell is blocked now; the other four states still compute
(BUG-1390).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HwMNZEzdzmrbPhuRNmZe3C
Upstream e628118d, 7d545039, f75a6e77, and ac565836. Dignity follows the sign, Deeptadi keeps every condition that applies, and Shayanadi hours come from the previous Hindu sunrise. BUG-1389.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream 9f9f16b3 and 872eaecf. Saham, Gulika, and Shayanadi opt into Hindu rising. The default window used by rectification and Panchanga stays the apparent-disk one. BUG-1388.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream ce9fed15. The exception now tests C on the forward arc from B to A, in both the value and the trace. BUG-1387.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Move the bde5d615 engine and full-synastry skill in unchanged, save both profiles without billing or a page, and retire the old synastry module.
Checked targeted pytest, tsc 0, lint 0 errors, service tests 6/6, and test:db 73 pass / 13 fail matching the Windows known list.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream e7301feb, 86935c0a, and 4f1536f2. Ketu reverses the count, the ninth is the fifth's obstruction, and a remaining tie stays strength_comparison_required so the consultation card shows 待定. BUG-1386.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream a4d4a249. Dig tables, Capricorn's first half, the sputa formula, and the benefic set now match jyotishganit_mirrored_equal_house_v2. Mercury is neutral. The result names that profile and does not call it a classical ruling. BUG-1385.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream 30b43643. Yoga Pinda multiplied only the bindu in the planet's own sign. It now uses the Trikona and Ekadhipatya rows, with PVR table 28 including Virgo 6. PVR example 43 is 152. BUG-1384.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream 8eec9a54. Floating accumulation and a 120-degree remainder put star, pada, and sub-lord on different sides of the same boundary. Random longitudes away from those boundaries are unchanged. BUG-1383.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream 3c2dad95, rewritten for this tree. A birth-clipped antardasha start was reported as a transition. The first mahadasha start stays, because BUG-1268 already stores the real pre-birth boundary there. BUG-1382.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream 3c2dad95, 8643bc9a, and the lon_to_nakshatra part of 9e04f0db. Exact star starts were split across int-division and modulo, so 16 of 27 openings reported a zero balance and skipped the mahadasha. Off-boundary longitudes keep the old modulo. BUG-1381.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Upstream 8643bc9a. The nakshatra breakdown was restarting the nine antardashas from the birth remainder, so mid-Ashwini reported Ketu instead of Rahu. Reuse cmd_dasha and clip to birth. BUG-1380.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Expired daily files are pruned before a new backup, a verify older than 36 hours raises an alert, and the host monitors call sudo with the production env file. The Linux 0600 check is on the staging quick list.
Co-Authored-By: Grok (agent loop)
Safety-net records move to BUG-1358 through BUG-1361 because staging already uses BUG-1355 through BUG-1357 for capacity.
Co-Authored-By: Grok (agent loop)
Product chose to keep the scheduled backup, but the current recovery point must stay the one created by hand. Daily runs still encrypt and restore-verify, then record a separate latest-backup note.
Co-Authored-By: Grok (agent loop)
The fix brief is already on staging. BUG-1354 on this branch collided with an earlier answer-effort record, so the safety-net entries move to BUG-1355 through BUG-1357.
Co-Authored-By: Grok (agent loop)
Consult, rectification turns, and report generation stop before a credit reservation when the product caps are hit. A full heavy-compute gate waits up to 8 seconds, and one consultation retries that 429 inside the same request.
Co-Authored-By: Grok (agent loop)
The bounced contract still looked for the old chart calls. The research script now writes to a temp directory unless --out is set. Quota splits stay unimplemented.
Co-Authored-By: Grok (agent loop)
The chart API rebuilt the ascendant without nakshatra, lord, and pada, so those cells stayed blank. Cache schema is now 7. Centered column headers line up with their cells, and the current dasha card keeps its tint without the left accent.
pytest 34 passed. tsc 0 errors. lint 0 errors and 123 existing warnings.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Synastry charts now skip the unused external overview and run together. Reports skip PyJHora subprocesses when that library is missing, and a full heavy-compute slot logs which paths hold it. Splitting quotas stays a simulation.
Gate, growth, and report pytest passed, as did the frontend route tests, tsc, lint, and the A/B script.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
Host rehearsal no longer needs Node on the production machine, and a failed cleanup now stops with the rehearsal database name. Docs-only staging heads no longer block rollback. Tests no longer hardcode Windows paths, and a trailing slash on the production site still refuses the test code.
pytest 12 passed; tsc exit 0; lint 0 errors; identity tests 9 passed, 0 cancelled. test:db was not run.
Co-Authored-By: Grok (agent loop)
The audit scans frontend app/components source. Its only Case Validation
hit was evidence-audit-panel.tsx, never mounted and deleted in #63, so the
topic is truthfully gone and case_validator's UX level is thin. Product
owner chose option A on issue #63 (Claude fixes directly). Staging gate
run 1824 was red on this test.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LutgzyJqNhPpZy1nFsJtN3
Cache work that depends only on the candidate minute, drop unread response fields, and reject windows above 241 candidates before any chart is built. Eleven canonical scenario hashes match 77057a819. Sealed and reported-offset replays match the 2026-10-08 records (changed_trial_count 8 and 581) and are stored as scoring_perf_2026_10_13.
Co-Authored-By: Grok (agent loop)
