Merge remote-tracking branch 'origin/staging' into codex/login-rectify-loading-1103

# Conflicts:
#	docs/BUG_HISTORY.md

CHANGELOG.md

@@ -1,5 +1,9 @@
 # 印度占星 Skill 更新日志
 
+## 2026-11-04 — Shadbala：月亮的盈亏力不再重复计入
+
+- 六力里月亮的盈亏力（Paksha）以前被算了四倍（时力里乘 2，动力又取乘 2 后的值），现按 BPHS（Santhanam 译本）27.10–11 + 27.18：时力里不乘 2，动力等于盈亏力，合计两倍。月亮的六力总分、Ishta / Kashta 和「最强星」可能变化（例：奥巴马公开盘最强星由月亮变为太阳），其他六颗星不变；结果里 `method_variants.moon_paksha` 写明口径（BUG-1557）。生时校正结果不变。Skill 版本不变；没有数据库变更。
+
 ## 2026-11-04 — 对话数据卡：每个领域宫位带上 Ashtakavarga 强 / 中 / 弱
 
 - 对话回答读的数据卡里，每个领域列出的宫位（如父母的 4、9 宫，婚恋的 5、7 宫，事业的 10 宫）多一项 `sav_strength`：该宫星座的原始 SAV 写成「强 / 中 / 弱」，不写点数。分档与星盘页一致：≥ 28 强、25–27 中、≤ 24 弱（来源 `references/ashtakavarga-complete-system.md` §三，与引擎 `scripts/ashtakavarga.py` 同界）。以前这项只能靠模型额外查，回测里 233 份回答只查了 2 次。

docs/BUG_HISTORY.md

@@ -21283,7 +21283,7 @@
 - 根因：`phase = abs(Sun - Moon) / 3` 没有折回。Raman《Graha and Bhava Balas》：月日之差超过 180° 时用 360° 去减，余数除以 3 为吉星的 Paksha，凶星为 60 减之，月亮加倍。PyJHora 4.8.6 `_paksha_bala` 有同样写法，所以与 PyJHora 对照时这一项一直「一致」。
 - 修复：`paksha_phase` 先折回 0–180° 再除以 3。
 - 验证：`tests/test_shadbala_paksha_fold.py`（折回、对称、满月 60、原始差 340° 的盘木星 Paksha 恰为 6.67——旧公式为 113.33）；Shadbala 相关测试改前改后失败清单一致（10 条既有失败均为 oracle / 审计研究测试，如「已验证案例 ≥ 4」实为 2，与本单无关）。
-- 同轮核对结论（不改代码）：Kala 里某一颗星差 ±60 是 PyJHora 时主公式错（星期一日出第一时主给成火星）；金星、火星、木星 Chesta 大差中 PyJHora 是离群值（乔布斯盘火星：本仓 21.6、VedAstro 21.7、jyotishganit 22.9、PyJHora 98.6）；太阳、月亮 Chesta 取 Ayana / Paksha 是 Raman 口径，VedAstro、PyJHora 给 0，属流派差。亏月（满月到新月）按凶星取「60 减」使近新月的月亮 Paksha 接近满分；占星师 2026-11-04 回复：外部实现也有此口径（如 saravali.github.io bala_kala），不是公式错误，不把月亮在各模块统一改成吉星；若产品要采用「月亮盈亏力随亮度增减」（满月 60、新月 0、上下弦同夹角等值），只能单独改月亮的 Paksha，并绑定具名来源与计算版本（月亮是否乘 2、与 Chesta 如何配合随同一算法），不得无差别修补。当前首要问题是两条计算路径口径不一致，见 BUG-1548。四家引擎 Sthana、Dig、Drik 也普遍不一致（`three_engine_parity_replay_manifest.json` 已记 `blocked_cross_engine_method_conflict`），Shadbala 绝对值暂不作数据卡上的强弱依据。
+- 同轮核对结论（不改代码）：Kala 里某一颗星差 ±60 是 PyJHora 时主公式错（星期一日出第一时主给成火星）；金星、火星、木星 Chesta 大差中 PyJHora 是离群值（乔布斯盘火星：本仓 21.6、VedAstro 21.7、jyotishganit 22.9、PyJHora 98.6）；太阳、月亮 Chesta 取 Ayana / Paksha，VedAstro、PyJHora 给 0，属流派差（更正，见 BUG-1557：原写「是 Raman 口径」不准确。Raman《Graha and Bhava Balas》例 56 的 Shadbala 总分里太阳、月亮没有 Chesta；「月亮 Chesta = Paksha」出自 BPHS 27.18）。亏月（满月到新月）按凶星取「60 减」使近新月的月亮 Paksha 接近满分；占星师 2026-11-04 回复：外部实现也有此口径（如 saravali.github.io bala_kala），不是公式错误，不把月亮在各模块统一改成吉星；若产品要采用「月亮盈亏力随亮度增减」（满月 60、新月 0、上下弦同夹角等值），只能单独改月亮的 Paksha，并绑定具名来源与计算版本（月亮是否乘 2、与 Chesta 如何配合随同一算法），不得无差别修补。当前首要问题是两条计算路径口径不一致，见 BUG-1548。四家引擎 Sthana、Dig、Drik 也普遍不一致（`three_engine_parity_replay_manifest.json` 已记 `blocked_cross_engine_method_conflict`），Shadbala 绝对值暂不作数据卡上的强弱依据。
 - 防复发：回归测试；对照时「两家一致」不等于正确，盈亏力这类两家同源写法要对照原书。
 - 关联记录：BUG-1188、BUG-1189、BUG-1270。
 - 复发自：无
@@ -21335,7 +21335,23 @@
 - 复发自：无
 - 修复版本：分支 `codex/vimshottari-second-cycle-20261104`（未推送）。
 
-## BUG-1557 | 咨询回答一上来叫用户名字；关键结论和依据没有任何强调；用户要「三条习惯」时写成一段「一是……二是……三是……」
+## BUG-1557 | Shadbala 月亮盈亏力（Paksha）计入四倍：Kala 里乘 2，Chesta 又取乘 2 后的值
+
+- 状态：resolved（引擎 + 回归测试；未部署）
+- 首次发现 / 最近更新：2026-11-04（BUG-1546 之后的 Paksha 口径研究：逐一对照 BPHS 梵文与 Santhanam 英译、Raman GBB、PVR、PyJHora、VedAstro、jyotishganit、saravali、Maitreya8、jyotishyamitra）/ 2026-11-04
+- 影响面：`scripts/shadbala.py` 精确路径 `calc_kala_bala_precise` 的月亮 paksha，连带 `calc_chesta_bala_precise` 的月亮 Chesta、月亮 Ishta / Kashta、Kala 与 Shadbala 总分、排名和「最强星」。凡走精确路径的都受影响：`cmd_shadbala`、全读、报告、星盘页六力表与 remedies（BUG-1548 以后同一路径）、Shadbala 服务接口。其他六颗星不变；`classify_drik_planets`（Drik 分类）不变；旧的简化路径 `calc_kala_bala` / `calc_chesta_bala` 未动（只有生时校正调用，只取 Sthana / Drik / Naisargika）。
+- 现象：乔布斯公开盘月亮 Kala·paksha 21.34、Chesta 21.34；奥巴马公开盘月亮 Chesta 73.88（超过 60），Ishta 64.99（超过 60），Kashta 被截成 0。月亮来自盈亏力的总分是基数 P 的 4 倍。
+- 触发条件：任何一张盘走精确路径。
+- 根因：精确路径同时用了两家互斥的写法——Raman《Graha and Bhava Balas》§55(c)「Chandra's Paksha Bala is always to be doubled」（例 56 的总分里月亮 Chesta 为空），和 BPHS 27.18「Candr's Paksh Bal will itself be her Chesht Bal」（27.10–11 不乘 2）；而且 Chesta 拿的是已经乘 2 的值。调查的所有来源里月亮盈亏力在总分中都是 2P，只有本仓是 4P。BUG-1546 和代码注释说「月亮 Chesta 取 Paksha 是 Raman 口径」，不准确（已在 BUG-1546 更正）。
+- 修复：产品 2026-11-04 采用候选 B「BPHS-Santhanam 27.10–11 + 27.18」（与占星师引用的 saravali.github.io、Maitreya8 默认设置、jyotishyamitra 同口径）。月亮 Paksha = P，基数为折回角 min(Δ, 360 − Δ) / 3，不乘 2；上弦（盈）月按吉星取 P，下弦（亏）月按凶星取 60 − P——`classify_drik_planets` 本来就按 Δ ≤ 180° 归吉、否则归凶，与 BPHS 3.11（Santhanam：decreasing Candr 为凶）一致，未改；月亮 Chesta = Paksha（BPHS 27.18），总分里共 2P。`calc_shadbala` 的 `method_variants` 新增 `moon_paksha: "bphs_santhanam_27_10_11_chesta_27_18"`（常量 `MOON_PAKSHA_CONVENTION`，注释写出处）；简化路径为 `legacy_approximation`。
+- 验证：`tests/test_shadbala_moon_paksha_convention.py`（乔布斯、梦露、奥巴马三张公开盘：月亮 Paksha = P 不是 2P、Chesta = Paksha、Chesta / Ishta / Kashta 都在 0–60、吉凶侧与 BPHS 3.11 一致、方法标签；乔布斯 Paksha 10.67、总分 638.06；奥巴马 Kashta 不再为 0；其他六星 Paksha 不变）。按「原值 / 新值 / 原因」更新：`test_shadbala_precise_context.py`（月亮 Kala 对存档 PyJHora 的差 52.17 → 41.51，PyJHora 仍乘 2；月亮 Chesta = 10.66）、`test_shadbala_paksha_fold.py`（月亮上限 120 → 60）、`test_shadbala_service_api.py`（method_variants 加 moon_paksha）、`test_vp_jain_shadbala_benchmark.py`（容差内 36 → 35、方法差 6 → 7：月亮 Kala 205.96 → 151.58，公布值 205.85 按 Raman 乘 2；月亮 Chesta 108.76 → 54.38，公布值 0；两项合计 205.96 对 205.85 + 0，仍差 0.11，新增方法差标签 `moon_paksha_convention`）。三张公开盘月亮总分（Raman 岁差，平均交点）：乔布斯 10.99 → 10.63 Rupa（仍最强）、梦露 9.04 → 8.28（仍第 2）、奥巴马 10.14 → 8.91（第 1 → 第 2，最强星由月亮变太阳）。Shadbala 相关 123 个测试文件改前改后失败清单一致（22 条既有失败，其中 `test_shadbala_*` 10 条，均为 oracle / 审计研究测试）。
+- 冻结身份（ERR-110）：`shadbala.py` 在生时校正冻结身份里。按 ERR-110 改 FREEZE / REPORT 常量后用各脚本 `--freeze` 新建记录并回放，标签 `moon_paksha_2026_11_04`（`docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.{freeze.json,json}`、`docs/research/reported_offset_moon_paksha_2026_11_04.{freeze.json,json}`），合同 `references/rectification_sealed_holdout.v1.json` 按 11-04 data_audit 先例轮换，旧记录逐字节保留。与 `data_audit_2026_11_04` 逐条逐字段比：密封 20 条全部相同、`metrics` 不变；报时窗口 900 条全部相同、`summary` 不变；相对 09-20 历史报告的 changed_trial_count 仍为 8 / 581。符合预期：校正打分只读 Sthana / Drik / Naisargika，不读 Kala / Chesta，Drik 分类未动。
+- 防复发：回归测试锁「月亮 Chesta = Paksha、Paksha 不乘 2、Ishta / Kashta 在 0–60」；口径用具名常量绑定并写进结果的 `method_variants`；对照外部引擎时按分量拆开比，Kala 与 Chesta 分别对上不等于总分口径一致。
+- 关联记录：BUG-1546（折回；本条更正其「Raman 口径」一句）、BUG-1548（星盘页与解读同一路径）。
+- 复发自：无
+- 修复版本：分支 `codex/moon-paksha-bphs-20261104`（未推送）。
+
+## BUG-1558 | 咨询回答一上来叫用户名字；关键结论和依据没有任何强调；用户要「三条习惯」时写成一段「一是……二是……三是……」
 
 - 状态：investigating（提示词 + 样式 + 合同测试；待 staging 真机同题对照，本机无模型密钥与排盘引擎，未跑回测）
 - 首次发现 / 最近更新：2026-11-04（产品负责人拿竞品同题「借贷 + 2/11/12 宫 + 三条理财习惯」对照反馈）/ 2026-11-04

docs/research/pre_work_error_ledger.md

@@ -17,6 +17,7 @@ Purpose: read this file before substantial project work. It exists to stop repea
 - BUG-1545 / BUG-1546 改了冻结身份里的 `narayana_dasha.py`、`active_rectification_event_engine.py`、`shadbala.py`，合入上游同步 10 后快速门 `test_rectification_validation_integrity_gate` 4 条红（ERR-110 同型）。按 ERR-110 重冻为 `data_audit_2026_11_04`：密封 20 条与 `d60_mapping_2026_10_29` 全同；报时 900 条 2 条 `true_rank` 变（Narayana 第二轮，同一例半径 60），`summary` 不变。教训：「只影响第一轮之后的年龄」不等于回放不变，重冻后一律逐条比，有变化照实写。
 - BUG-1556（worktree `/workspace/wt-vim`，分支 `codex/vimshottari-second-cycle-20261104`）：Vimshottari 过 120 年变空已修，改在冻结身份之外（`chart_dasha/birth_moment.py` 的 `next_cycle_mahadashas`，`cmd_dasha` 与星盘 / 咨询共用），`dasha_analyzer.py`、`domain_calculation_service.py` 未动，无需重冻。`pre_work_check.py` exit 0；Python quick 门一次通过。注意：`_compute_chart_sync` 里 Vimshottari 算了两次，后一次（原 `utcnow()`）才是返回的 `dasha`，改参照日要两处都改；`test_chart_dasha_birth_moment_20261028.py` 里有 1879 年等老盘，以前只因两边都只排一轮才相等。
 - 数据卡宫行加 Ashtakavarga 强 / 中 / 弱（worktree `/workspace/wt-card-av`，`codex/card-ashtakavarga-20261104`）：`pre_work_check.py` 第一次 exit 1（`fragment_scan` 90 s、`focused_tests` 45 s 超时，当时同机在建 worktree、跑卡片导出），机器空下来重跑 pass、六项全部 ok。分档沿用星盘页 `savBand`（≥ 28 强、25–27 中、≤ 24 弱），来源写在 `SAV_BAND_SOURCE`。注意：事业卡的 `sav_by_house`、财富 / 综合卡的 `ashtakavarga` 块本来就把原始与净化后的 SAV 点数给了模型，本单没动。
+- BUG-1557 月亮 Paksha 口径（worktree `/workspace/wt-moon`，`codex/moon-paksha-bphs-20261104`）：`pre_work_check.py` exit 0、六项全部 ok。精确路径 Kala 里月亮 ×2（Raman）加 Chesta = Paksha（BPHS 27.18）同时用、Chesta 又取 ×2 后的值，总分 4P；改为 BPHS-Santhanam（不乘 2，Chesta = Paksha），`method_variants.moon_paksha` 标口径。`shadbala.py` 在冻结身份里，按 ERR-110 重冻为 `moon_paksha_2026_11_04`：密封 20 条、报时 900 条与 `data_audit_2026_11_04` 全同。教训：两家来源各自「×2」「Chesta = Paksha」都对，拼在一起就错；引进一条古典规则时先看同一来源怎么处理相邻分量。Shadbala 相关测试改前改后失败清单一致（同上 10 条 + 其他既有 12 条）；Python quick 门一次通过。
 
 ## 2026-11-03 · 输入框单行起步（方案 2，worktree `/workspace/wt-composer`）
 

docs/research/reported_offset_moon_paksha_2026_11_04.freeze.json

@@ -0,0 +1,118 @@
+{
+  "record_version": "reported-offset-native-candidate-date-v3",
+  "frozen_at_utc": "2026-11-04T07:22:52.617248+00:00",
+  "ayanamsa": "raman",
+  "node_mode": "mean",
+  "radii_minutes": [
+    15,
+    30,
+    60
+  ],
+  "offsets_minutes": [
+    -30,
+    -20,
+    -15,
+    -10,
+    -8,
+    -5,
+    -3,
+    0,
+    3,
+    5,
+    8,
+    10,
+    15,
+    20,
+    30
+  ],
+  "minute_step": 1,
+  "dataset": "references/real_case_calibration/minute_rectification_holdout_v3.json",
+  "dataset_sha256": "45aa76ab79dfc6c6404b6252083e1aa5c7e2e2b7de811fc11042a8ad2da3f0ea",
+  "dataset_benchmark_id": "minute_rectification_fact_ranker_v4_holdout_v3",
+  "implementation_sha256": "19ce36a9938928f345928fc44792eb70e6b71945aa5ab2a08a942b214cf0b085",
+  "implementation_sha256_prefix": "19ce36a9938928f3",
+  "historical_artifacts_manifest_sha256": "6102a26a840be207b5858b3a4c0509274468ae9071e86d308cbbac7371d6864e",
+  "production_scoring_files": [
+    "scripts/active_rectification_event_engine.py",
+    "scripts/active_rectification_events.py",
+    "scripts/ashtakavarga.py",
+    "scripts/dasha_analyzer.py",
+    "scripts/divisional_charts_extended.py",
+    "scripts/domain_calculation_service.py",
+    "scripts/functional_benefics.py",
+    "scripts/jaimini.py",
+    "scripts/minute_rectification_fact_ranker_v4.py",
+    "scripts/minute_rectification_feature_facts_v4.py",
+    "scripts/narayana_dasha.py",
+    "scripts/rectification/api_service.py",
+    "scripts/rectification/candidate_contrast.py",
+    "scripts/rectification/candidate_window.py",
+    "scripts/rectification/case_holdout.py",
+    "scripts/rectification/contracts.py",
+    "scripts/rectification/dasha_transition_proximity.py",
+    "scripts/rectification/decision_policy.py",
+    "scripts/rectification/event_probes.py",
+    "scripts/rectification/refinement_packet.py",
+    "scripts/rectification/scoring_service.py",
+    "scripts/shadbala.py",
+    "scripts/varga.py"
+  ],
+  "production_scoring_sha256": "9857465d8305d0ccdcedff12c4c0bc572d4a12eea7e348fa4fc1b38e4f244fcf",
+  "research_files": [
+    "scripts/research/reported_offset_sweep.py",
+    "scripts/research/cluster_width_lib.py",
+    "scripts/research/probe_supply_after_six.py",
+    "scripts/research/sealed_holdout_rerun.py",
+    "scripts/minute_rectification_blind_eval.py",
+    "scripts/minute_rectification_fact_blind_eval_v4.py",
+    "scripts/minute_rectification_holdout_validator.py"
+  ],
+  "research_implementation_sha256": "8747a78a8b7a3fee0542423978adab11cee3e98880065ce13ef243f9488e5de9",
+  "file_sha256": {
+    "scripts/active_rectification_event_engine.py": "dced0b398ad90bdec59c743a55ffe01613306127c17a1e44c8c9cfb569c57841",
+    "scripts/active_rectification_events.py": "03e333ccf1103caf685dfe50fe890ab1793e945037085f50ce834a4a3b232222",
+    "scripts/ashtakavarga.py": "c524da3f54b2ed9c080b78b33854854a0ca3d5d79b7e536bb1c3bc2be3e02852",
+    "scripts/dasha_analyzer.py": "3949b96813217fe624899856afec9f0b787c4cb0fea3a9a0d961990dbd11943a",
+    "scripts/divisional_charts_extended.py": "e9820e80f14fddc86ab5495d11455d16c2d3c9502a9a242fcf9efc41186e0d11",
+    "scripts/domain_calculation_service.py": "66923a2bc69daf65c965d2973c4b9234edbec5147909eaf3609b6b377ce3c7ca",
+    "scripts/functional_benefics.py": "cbaac3b8c09b97f8724b5300e9a402cbf77089af6c7db8788550d40fd35ac57b",
+    "scripts/jaimini.py": "6b88658da5404da9b136686d92a04be462d61a39da6e8de4255021e2ed2d1173",
+    "scripts/minute_rectification_blind_eval.py": "07aa30d8bad3278669495d5fea8be937aaf6367c8af980ff4f2fe9aa84ccc6ff",
+    "scripts/minute_rectification_fact_blind_eval_v4.py": "bdf460e3467208e8b87345c5beb2128f3e22de7eeffd68f72b281b3ad9affc92",
+    "scripts/minute_rectification_fact_ranker_v4.py": "f1a0d8d2482cd754d2e7f253a9606b0373519435a2bf58cfc3753e2b9ae0fb76",
+    "scripts/minute_rectification_feature_facts_v4.py": "3d899a2cdd4825b8e32062443d487edcd2173a4ea55a36eb9f3abe587c0faf0f",
+    "scripts/minute_rectification_holdout_validator.py": "b60b1c752e58672beaee8b36eb3b2348169726cda070ec969db0f13e7ed715af",
+    "scripts/narayana_dasha.py": "9a847d3b8d05f90d4f275079c97af1f71e83a74f930e14bd8f3f37b293d91d1e",
+    "scripts/rectification/api_service.py": "089730097d3872a4f7b5d7dd5574bb7e9ab4d2e246d4a380800bc824d7979b95",
+    "scripts/rectification/candidate_contrast.py": "eee7da71c0574a08876404b78ca111153fa95fb83ace82b9fdca141d1992e273",
+    "scripts/rectification/candidate_window.py": "2fbc5058b12049b15fca375cdc125b8283e19f9f3e4bedcc291e344501891bef",
+    "scripts/rectification/case_holdout.py": "fe0e699b617d9e216acdfeff87af4221c1d6f4c101bf33e1dbf14e2bcd75a8b5",
+    "scripts/rectification/contracts.py": "c6dc6bd230e38f9f2126c224aae9ddbd6415304c15340f8c7cd3f06f6ef3160f",
+    "scripts/rectification/dasha_transition_proximity.py": "5ce4ed27b0f06c681abc34e5fd998fafb10c153a0070aa36c99ddfa869a2cbf2",
+    "scripts/rectification/decision_policy.py": "80e8de4680d19a052213bb821d2a9c94143a2962235aeb27f6ad81374d2e60d5",
+    "scripts/rectification/event_probes.py": "f5ca9d54501f67a1546fcf5b01da0c1277dec2e2c6df481884e8900d14909a7c",
+    "scripts/rectification/refinement_packet.py": "dbeea5aa780889785ce850f3881e8b7f67bb44594e9bc382c65b2e730884a2f4",
+    "scripts/rectification/scoring_service.py": "fa056fa30d9e6ffa4b75295fc49c793e35b26ebeca64ca843d26c8abef3050f2",
+    "scripts/research/cluster_width_lib.py": "1d8f2ffa7795d62069deebb2de8e16749317037d4ed3f24a2c0092676e5cfcc6",
+    "scripts/research/probe_supply_after_six.py": "098a5b4398d6f8d996917d809c2fc922ad64031d3b70660f8ef7c876aba92438",
+    "scripts/research/reported_offset_sweep.py": "1c7183e77f442b8c6b46761ff8a833b6c27b7841999426c72556237f45df1343",
+    "scripts/research/sealed_holdout_rerun.py": "957b824b7579c4f52681c85d7964fc802f8179470c93917cded63c7fe6c2769d",
+    "scripts/shadbala.py": "352a9dd91c9be6dd7a294b1dc24c62173e8f08db5fc36c453dc5dcd7e0fd62d7",
+    "scripts/varga.py": "1bca389e4f36d293a755de175163c3f39b42bab5831cb6f7828830899f9f177e"
+  },
+  "hash_scope": "explicit_identity_file_sets_not_a_transitive_dependency_lock",
+  "scorer": "native_event_contribution_matrix_not_shadow_fact_ranker",
+  "delivery": "initial_signature_clusters_peak_gap_lt_8_envelope_no_answers_no_elimination",
+  "rank": "score_desc_then_sha256(benchmark_id:case_id:candidate_time)",
+  "grid": "every_minute_inclusive_not_production_two_minute_sampling",
+  "cross_midnight": "date_aware_candidates_distance_and_native_single_matrix",
+  "evaluator_sha256": "1c7183e77f442b8c6b46761ff8a833b6c27b7841999426c72556237f45df1343",
+  "replay_revision": "native_candidate_date_v3",
+  "supersedes": "candidate_date_grouped_v2_identity_and_path_not_assumed_numerically_wrong",
+  "is_blind_evaluation": false,
+  "truth_hidden_from_ranker": true,
+  "official_valid_independent_blind": false,
+  "official_blind_trial_count": 0,
+  "results_previously_seen": true,
+  "must_not_use_for_tuning": true
+}

docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.freeze.json

@@ -0,0 +1,116 @@
+{
+  "record_version": "exposed-v3-fixed-protocol-cross-midnight-rerun-v2",
+  "extended_identity": {
+    "historical_artifacts_manifest_sha256": "6102a26a840be207b5858b3a4c0509274468ae9071e86d308cbbac7371d6864e",
+    "production_scoring_files": [
+      "scripts/active_rectification_event_engine.py",
+      "scripts/active_rectification_events.py",
+      "scripts/ashtakavarga.py",
+      "scripts/dasha_analyzer.py",
+      "scripts/divisional_charts_extended.py",
+      "scripts/domain_calculation_service.py",
+      "scripts/functional_benefics.py",
+      "scripts/jaimini.py",
+      "scripts/minute_rectification_fact_ranker_v4.py",
+      "scripts/minute_rectification_feature_facts_v4.py",
+      "scripts/narayana_dasha.py",
+      "scripts/rectification/api_service.py",
+      "scripts/rectification/candidate_contrast.py",
+      "scripts/rectification/candidate_window.py",
+      "scripts/rectification/case_holdout.py",
+      "scripts/rectification/contracts.py",
+      "scripts/rectification/dasha_transition_proximity.py",
+      "scripts/rectification/decision_policy.py",
+      "scripts/rectification/event_probes.py",
+      "scripts/rectification/refinement_packet.py",
+      "scripts/rectification/scoring_service.py",
+      "scripts/shadbala.py",
+      "scripts/varga.py"
+    ],
+    "production_scoring_sha256": "9857465d8305d0ccdcedff12c4c0bc572d4a12eea7e348fa4fc1b38e4f244fcf",
+    "research_files": [
+      "scripts/research/reported_offset_sweep.py",
+      "scripts/research/cluster_width_lib.py",
+      "scripts/research/probe_supply_after_six.py",
+      "scripts/research/sealed_holdout_rerun.py",
+      "scripts/minute_rectification_blind_eval.py",
+      "scripts/minute_rectification_fact_blind_eval_v4.py",
+      "scripts/minute_rectification_holdout_validator.py"
+    ],
+    "research_implementation_sha256": "8747a78a8b7a3fee0542423978adab11cee3e98880065ce13ef243f9488e5de9",
+    "file_sha256": {
+      "scripts/active_rectification_event_engine.py": "dced0b398ad90bdec59c743a55ffe01613306127c17a1e44c8c9cfb569c57841",
+      "scripts/active_rectification_events.py": "03e333ccf1103caf685dfe50fe890ab1793e945037085f50ce834a4a3b232222",
+      "scripts/ashtakavarga.py": "c524da3f54b2ed9c080b78b33854854a0ca3d5d79b7e536bb1c3bc2be3e02852",
+      "scripts/dasha_analyzer.py": "3949b96813217fe624899856afec9f0b787c4cb0fea3a9a0d961990dbd11943a",
+      "scripts/divisional_charts_extended.py": "e9820e80f14fddc86ab5495d11455d16c2d3c9502a9a242fcf9efc41186e0d11",
+      "scripts/domain_calculation_service.py": "66923a2bc69daf65c965d2973c4b9234edbec5147909eaf3609b6b377ce3c7ca",
+      "scripts/functional_benefics.py": "cbaac3b8c09b97f8724b5300e9a402cbf77089af6c7db8788550d40fd35ac57b",
+      "scripts/jaimini.py": "6b88658da5404da9b136686d92a04be462d61a39da6e8de4255021e2ed2d1173",
+      "scripts/minute_rectification_blind_eval.py": "07aa30d8bad3278669495d5fea8be937aaf6367c8af980ff4f2fe9aa84ccc6ff",
+      "scripts/minute_rectification_fact_blind_eval_v4.py": "bdf460e3467208e8b87345c5beb2128f3e22de7eeffd68f72b281b3ad9affc92",
+      "scripts/minute_rectification_fact_ranker_v4.py": "f1a0d8d2482cd754d2e7f253a9606b0373519435a2bf58cfc3753e2b9ae0fb76",
+      "scripts/minute_rectification_feature_facts_v4.py": "3d899a2cdd4825b8e32062443d487edcd2173a4ea55a36eb9f3abe587c0faf0f",
+      "scripts/minute_rectification_holdout_validator.py": "b60b1c752e58672beaee8b36eb3b2348169726cda070ec969db0f13e7ed715af",
+      "scripts/narayana_dasha.py": "9a847d3b8d05f90d4f275079c97af1f71e83a74f930e14bd8f3f37b293d91d1e",
+      "scripts/rectification/api_service.py": "089730097d3872a4f7b5d7dd5574bb7e9ab4d2e246d4a380800bc824d7979b95",
+      "scripts/rectification/candidate_contrast.py": "eee7da71c0574a08876404b78ca111153fa95fb83ace82b9fdca141d1992e273",
+      "scripts/rectification/candidate_window.py": "2fbc5058b12049b15fca375cdc125b8283e19f9f3e4bedcc291e344501891bef",
+      "scripts/rectification/case_holdout.py": "fe0e699b617d9e216acdfeff87af4221c1d6f4c101bf33e1dbf14e2bcd75a8b5",
+      "scripts/rectification/contracts.py": "c6dc6bd230e38f9f2126c224aae9ddbd6415304c15340f8c7cd3f06f6ef3160f",
+      "scripts/rectification/dasha_transition_proximity.py": "5ce4ed27b0f06c681abc34e5fd998fafb10c153a0070aa36c99ddfa869a2cbf2",
+      "scripts/rectification/decision_policy.py": "80e8de4680d19a052213bb821d2a9c94143a2962235aeb27f6ad81374d2e60d5",
+      "scripts/rectification/event_probes.py": "f5ca9d54501f67a1546fcf5b01da0c1277dec2e2c6df481884e8900d14909a7c",
+      "scripts/rectification/refinement_packet.py": "dbeea5aa780889785ce850f3881e8b7f67bb44594e9bc382c65b2e730884a2f4",
+      "scripts/rectification/scoring_service.py": "fa056fa30d9e6ffa4b75295fc49c793e35b26ebeca64ca843d26c8abef3050f2",
+      "scripts/research/cluster_width_lib.py": "1d8f2ffa7795d62069deebb2de8e16749317037d4ed3f24a2c0092676e5cfcc6",
+      "scripts/research/probe_supply_after_six.py": "098a5b4398d6f8d996917d809c2fc922ad64031d3b70660f8ef7c876aba92438",
+      "scripts/research/reported_offset_sweep.py": "1c7183e77f442b8c6b46761ff8a833b6c27b7841999426c72556237f45df1343",
+      "scripts/research/sealed_holdout_rerun.py": "957b824b7579c4f52681c85d7964fc802f8179470c93917cded63c7fe6c2769d",
+      "scripts/shadbala.py": "352a9dd91c9be6dd7a294b1dc24c62173e8f08db5fc36c453dc5dcd7e0fd62d7",
+      "scripts/varga.py": "1bca389e4f36d293a755de175163c3f39b42bab5831cb6f7828830899f9f177e"
+    },
+    "hash_scope": "explicit_identity_file_sets_not_a_transitive_dependency_lock"
+  },
+  "production_identity_scope": "context_only_shadow_scorer_does_not_call_transition_proximity",
+  "frozen_at_utc": "2026-11-04T07:22:50.077499+00:00",
+  "dataset_path": "references/real_case_calibration/minute_rectification_holdout_v3.json",
+  "dataset_sha256": "45aa76ab79dfc6c6404b6252083e1aa5c7e2e2b7de811fc11042a8ad2da3f0ea",
+  "algorithm_version": "birth-time-event-fact-ranker-v4-shadow",
+  "implementation_sha256": "19ce36a9938928f345928fc44792eb70e6b71945aa5ab2a08a942b214cf0b085",
+  "files": [
+    "scripts/active_rectification_event_engine.py",
+    "scripts/active_rectification_events.py",
+    "scripts/ashtakavarga.py",
+    "scripts/dasha_analyzer.py",
+    "scripts/divisional_charts_extended.py",
+    "scripts/domain_calculation_service.py",
+    "scripts/jaimini.py",
+    "scripts/minute_rectification_fact_ranker_v4.py",
+    "scripts/minute_rectification_feature_facts_v4.py",
+    "scripts/narayana_dasha.py",
+    "scripts/shadbala.py",
+    "scripts/varga.py"
+  ],
+  "historical_frozen_sha256": "f41c298dd6cdcebe7a93e632f7191954be7987f6012ca2a34cecb6e447fbf196",
+  "evaluator_sha256": "957b824b7579c4f52681c85d7964fc802f8179470c93917cded63c7fe6c2769d",
+  "ayanamsa": "raman",
+  "node_mode": "mean",
+  "candidate_radius_minutes": [
+    10
+  ],
+  "minute_step": 1,
+  "release_metrics": {
+    "top_1_rate_minimum": 0.6,
+    "top_3_rate_minimum": 0.85,
+    "mean_absolute_minute_error_maximum": 2.0,
+    "false_confirmation_rate_maximum": 0.05,
+    "correct_insufficient_evidence_rejection_rate_minimum": 0.9
+  },
+  "results_previously_seen": true,
+  "official_valid_independent_blind": false,
+  "must_not_use_for_tuning": true,
+  "tie_breaker": "sha256(benchmark_id:case_id:candidate_time); published minute is never passed to the ranker",
+  "metric_rank_definition": "competition_rank_1_plus_strictly_higher_scores_legacy_protocol",
+  "extra_metric_rank_definition": "score_desc_then_existing_opaque_sha256_total_order"
+}

frontend/src/app/globals.css

@@ -1763,7 +1763,7 @@ button.chat-chart-resize-handle:active { position: absolute; top: 0; left: -3px;
 .message-markdown h2, .message-markdown h3 { margin: 24px 0 10px; color: var(--color-ink); font-family: var(--font-display); font-weight: 500; letter-spacing: -.3px; }
 .message-markdown h2 { font-size: var(--type-display-sm); }
 .message-markdown h3 { font-size: var(--type-title-lg); }
-/* BUG-1557: 600, not 500 — at 500 the emphasis the answer now carries barely stood off the body. */
+/* BUG-1558: 600, not 500 — at 500 the emphasis the answer now carries barely stood off the body. */
 .message-markdown strong { color: var(--color-ink); font-weight: 600; }
 .message-markdown a { text-underline-offset: 3px; color: var(--color-action); }
 .message-markdown code { padding: 2px 5px; font-size: .9em; border-radius: var(--radius-xs); background: var(--color-canvas-muted); font-family: var(--font-mono); }

frontend/tests/consult-voice-emphasis-20261104.test.ts

@@ -4,7 +4,7 @@ import test from "node:test";
 
 import { productConversationVoice } from "../src/mastra/product-voice.ts";
 
-// BUG-1557: answers opened with the user's name, had no emphasis, and wrote
+// BUG-1558: answers opened with the user's name, had no emphasis, and wrote
 // three asked-for habits inline as 一是……二是……三是…….
 
 test("the answer never addresses the user by name", () => {

frontend/tests/consultation-thinking-plan.test.ts

@@ -179,7 +179,7 @@ test("natal spoken answers default to paragraphs, not bold-label lists (BUG-1244
   assert.match(productConversationVoice, /写成段落，段落之间空一行/);
   // 原值(3): /有三个以上候选时间窗时才用列表/
   // 新值(3): 用户要几条或有三个以上候选时间窗时用列表，每条带「为什么适合你」
-  // 原因(3): BUG-1557 产品对照竞品：用户说「给三条」时我们写成了一段「一是……二是……三是……」
+  // 原因(3): BUG-1558 产品对照竞品：用户说「给三条」时我们写成了一段「一是……二是……三是……」
   assert.match(productConversationVoice, /用户要几条（「给三条」）、或有三个以上候选时间窗时用列表，每条先一句做法、再一句为什么适合你/);
   assert.match(consultationSpokenHeadingRule("daily"), /Markdown bullet lists/);
   assert.doesNotMatch(consultationSpokenHeadingRule("window"), /start with ## /);

frontend/tests/rectification-confirmation-gate.test.ts

@@ -120,8 +120,11 @@ const diagnosticReport = JSON.parse(readFileSync(
 // 原值(13)：docs/research/sealed_holdout_rerun_d60_mapping_2026_10_29.json
 // 新值(13)：docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json
 // 原因(13)：数据核对 BUG-1545（narayana_dasha.py、active_rectification_event_engine.py）与 BUG-1546（shadbala.py）改了冻结身份文件，20 条回放不变，ERR-110 另存新记录。确认门行为与阈值不变。
+// 原值(14)：docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json
+// 新值(14)：docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.json
+// 原因(14)：BUG-1557 月亮 Paksha 口径改了冻结身份里的 shadbala.py，20 条回放不变，ERR-110 另存新记录。确认门行为与阈值不变。
 const currentTreeRerun = JSON.parse(readFileSync(
-  new URL("../../docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json", import.meta.url),
+  new URL("../../docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.json", import.meta.url),
   "utf8",
 )) as {
   frozen_record: { implementation_sha256: string };
@@ -272,9 +275,10 @@ test("sealed holdout aggregates match the v3 report and still block confirmation
   // 原值(10)：…upstream_sync8_2026_10_18.json / 新值(10)：…upstream_sync9b_2026_10_28.json / 原因(10)：sync9b 改了冻结身份里的 varga.py，输出不变，ERR-110 另存新记录。
   // 原值(11)：…upstream_sync9b_2026_10_28.json / 新值(11)：…d60_mapping_2026_10_29.json / 原因(11)：BUG-1500 修正 D60 映射，输出不变，ERR-110 另存新记录。
   // 原值(12)：…d60_mapping_2026_10_29.json / 新值(12)：…data_audit_2026_11_04.json / 原因(12)：BUG-1545 / BUG-1546 改了冻结身份文件，输出不变，ERR-110 另存新记录。
+  // 原值(13)：…data_audit_2026_11_04.json / 新值(13)：…moon_paksha_2026_11_04.json / 原因(13)：BUG-1557 改了冻结身份里的 shadbala.py，输出不变，ERR-110 另存新记录。
   assert.equal(
     productHoldout.current_tree_scorer.source_report,
-    "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json",
+    "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.json",
   );
   assert.equal(currentTreeRerun.official_valid_independent_blind, false);
   assert.equal(productHoldout.current_tree_scorer.matches_metrics_scorer, false);

references/rectification_sealed_holdout.v1.json

@@ -24,7 +24,7 @@
     "evaluated_on": "2026-07-21"
   },
   "current_tree_scorer": {
-    "implementation_sha256": "1de938b696d9e9639f4f401a83c3c51e231f279784335869471237b0f9101d7a",
+    "implementation_sha256": "19ce36a9938928f345928fc44792eb70e6b71945aa5ab2a08a942b214cf0b085",
     "extended_identity": {
       "historical_artifacts_manifest_sha256": "6102a26a840be207b5858b3a4c0509274468ae9071e86d308cbbac7371d6864e",
       "production_scoring_files": [
@@ -52,7 +52,7 @@
         "scripts/shadbala.py",
         "scripts/varga.py"
       ],
-      "production_scoring_sha256": "975b5755ea8ba7037baf1510015fcc57e64f0620ac866ff076685d317c6f5731",
+      "production_scoring_sha256": "9857465d8305d0ccdcedff12c4c0bc572d4a12eea7e348fa4fc1b38e4f244fcf",
       "research_files": [
         "scripts/research/reported_offset_sweep.py",
         "scripts/research/cluster_width_lib.py",
@@ -62,7 +62,7 @@
         "scripts/minute_rectification_fact_blind_eval_v4.py",
         "scripts/minute_rectification_holdout_validator.py"
       ],
-      "research_implementation_sha256": "e752fb86c6462b2d316902cba1e11fb55a630fe8abc43f871b53aa58a62422ed",
+      "research_implementation_sha256": "8747a78a8b7a3fee0542423978adab11cee3e98880065ce13ef243f9488e5de9",
       "file_sha256": {
         "scripts/active_rectification_event_engine.py": "dced0b398ad90bdec59c743a55ffe01613306127c17a1e44c8c9cfb569c57841",
         "scripts/active_rectification_events.py": "03e333ccf1103caf685dfe50fe890ab1793e945037085f50ce834a4a3b232222",
@@ -90,9 +90,9 @@
         "scripts/rectification/scoring_service.py": "fa056fa30d9e6ffa4b75295fc49c793e35b26ebeca64ca843d26c8abef3050f2",
         "scripts/research/cluster_width_lib.py": "1d8f2ffa7795d62069deebb2de8e16749317037d4ed3f24a2c0092676e5cfcc6",
         "scripts/research/probe_supply_after_six.py": "098a5b4398d6f8d996917d809c2fc922ad64031d3b70660f8ef7c876aba92438",
-        "scripts/research/reported_offset_sweep.py": "dfea4cd09d425fbe78b8b8e5de82eab4baaea41c1cd21ea470710e1dc5e405cd",
-        "scripts/research/sealed_holdout_rerun.py": "cbc1ec1fdf8e22c30df1d8bd9497b0b6274cc7d383b24fefbe432972746fac90",
-        "scripts/shadbala.py": "658f94eaef5ad5826d10c11b58cf1a2c95e502befaea095ce3f1079d6c2f0dec",
+        "scripts/research/reported_offset_sweep.py": "1c7183e77f442b8c6b46761ff8a833b6c27b7841999426c72556237f45df1343",
+        "scripts/research/sealed_holdout_rerun.py": "957b824b7579c4f52681c85d7964fc802f8179470c93917cded63c7fe6c2769d",
+        "scripts/shadbala.py": "352a9dd91c9be6dd7a294b1dc24c62173e8f08db5fc36c453dc5dcd7e0fd62d7",
         "scripts/varga.py": "1bca389e4f36d293a755de175163c3f39b42bab5831cb6f7828830899f9f177e"
       },
       "hash_scope": "explicit_identity_file_sets_not_a_transitive_dependency_lock"
@@ -100,7 +100,7 @@
     "matches_metrics_scorer": false,
     "official_eval_implementation_hash_matches": false,
     "official_eval_trial_count": 0,
-    "source_report": "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json",
+    "source_report": "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.json",
     "fixed_protocol_rerun_trial_count": 20,
     "fixed_protocol_rerun_hash_matches": true,
     "metrics": {
@@ -113,6 +113,106 @@
     }
   },
   "current_tree_fixed_protocol_rerun": {
+    "evaluated_on": "2026-11-04",
+    "report_path": "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.json",
+    "freeze_record_path": "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.freeze.json",
+    "implementation_sha256": "19ce36a9938928f345928fc44792eb70e6b71945aa5ab2a08a942b214cf0b085",
+    "extended_identity": {
+      "historical_artifacts_manifest_sha256": "6102a26a840be207b5858b3a4c0509274468ae9071e86d308cbbac7371d6864e",
+      "production_scoring_files": [
+        "scripts/active_rectification_event_engine.py",
+        "scripts/active_rectification_events.py",
+        "scripts/ashtakavarga.py",
+        "scripts/dasha_analyzer.py",
+        "scripts/divisional_charts_extended.py",
+        "scripts/domain_calculation_service.py",
+        "scripts/functional_benefics.py",
+        "scripts/jaimini.py",
+        "scripts/minute_rectification_fact_ranker_v4.py",
+        "scripts/minute_rectification_feature_facts_v4.py",
+        "scripts/narayana_dasha.py",
+        "scripts/rectification/api_service.py",
+        "scripts/rectification/candidate_contrast.py",
+        "scripts/rectification/candidate_window.py",
+        "scripts/rectification/case_holdout.py",
+        "scripts/rectification/contracts.py",
+        "scripts/rectification/dasha_transition_proximity.py",
+        "scripts/rectification/decision_policy.py",
+        "scripts/rectification/event_probes.py",
+        "scripts/rectification/refinement_packet.py",
+        "scripts/rectification/scoring_service.py",
+        "scripts/shadbala.py",
+        "scripts/varga.py"
+      ],
+      "production_scoring_sha256": "9857465d8305d0ccdcedff12c4c0bc572d4a12eea7e348fa4fc1b38e4f244fcf",
+      "research_files": [
+        "scripts/research/reported_offset_sweep.py",
+        "scripts/research/cluster_width_lib.py",
+        "scripts/research/probe_supply_after_six.py",
+        "scripts/research/sealed_holdout_rerun.py",
+        "scripts/minute_rectification_blind_eval.py",
+        "scripts/minute_rectification_fact_blind_eval_v4.py",
+        "scripts/minute_rectification_holdout_validator.py"
+      ],
+      "research_implementation_sha256": "8747a78a8b7a3fee0542423978adab11cee3e98880065ce13ef243f9488e5de9",
+      "file_sha256": {
+        "scripts/active_rectification_event_engine.py": "dced0b398ad90bdec59c743a55ffe01613306127c17a1e44c8c9cfb569c57841",
+        "scripts/active_rectification_events.py": "03e333ccf1103caf685dfe50fe890ab1793e945037085f50ce834a4a3b232222",
+        "scripts/ashtakavarga.py": "c524da3f54b2ed9c080b78b33854854a0ca3d5d79b7e536bb1c3bc2be3e02852",
+        "scripts/dasha_analyzer.py": "3949b96813217fe624899856afec9f0b787c4cb0fea3a9a0d961990dbd11943a",
+        "scripts/divisional_charts_extended.py": "e9820e80f14fddc86ab5495d11455d16c2d3c9502a9a242fcf9efc41186e0d11",
+        "scripts/domain_calculation_service.py": "66923a2bc69daf65c965d2973c4b9234edbec5147909eaf3609b6b377ce3c7ca",
+        "scripts/functional_benefics.py": "cbaac3b8c09b97f8724b5300e9a402cbf77089af6c7db8788550d40fd35ac57b",
+        "scripts/jaimini.py": "6b88658da5404da9b136686d92a04be462d61a39da6e8de4255021e2ed2d1173",
+        "scripts/minute_rectification_blind_eval.py": "07aa30d8bad3278669495d5fea8be937aaf6367c8af980ff4f2fe9aa84ccc6ff",
+        "scripts/minute_rectification_fact_blind_eval_v4.py": "bdf460e3467208e8b87345c5beb2128f3e22de7eeffd68f72b281b3ad9affc92",
+        "scripts/minute_rectification_fact_ranker_v4.py": "f1a0d8d2482cd754d2e7f253a9606b0373519435a2bf58cfc3753e2b9ae0fb76",
+        "scripts/minute_rectification_feature_facts_v4.py": "3d899a2cdd4825b8e32062443d487edcd2173a4ea55a36eb9f3abe587c0faf0f",
+        "scripts/minute_rectification_holdout_validator.py": "b60b1c752e58672beaee8b36eb3b2348169726cda070ec969db0f13e7ed715af",
+        "scripts/narayana_dasha.py": "9a847d3b8d05f90d4f275079c97af1f71e83a74f930e14bd8f3f37b293d91d1e",
+        "scripts/rectification/api_service.py": "089730097d3872a4f7b5d7dd5574bb7e9ab4d2e246d4a380800bc824d7979b95",
+        "scripts/rectification/candidate_contrast.py": "eee7da71c0574a08876404b78ca111153fa95fb83ace82b9fdca141d1992e273",
+        "scripts/rectification/candidate_window.py": "2fbc5058b12049b15fca375cdc125b8283e19f9f3e4bedcc291e344501891bef",
+        "scripts/rectification/case_holdout.py": "fe0e699b617d9e216acdfeff87af4221c1d6f4c101bf33e1dbf14e2bcd75a8b5",
+        "scripts/rectification/contracts.py": "c6dc6bd230e38f9f2126c224aae9ddbd6415304c15340f8c7cd3f06f6ef3160f",
+        "scripts/rectification/dasha_transition_proximity.py": "5ce4ed27b0f06c681abc34e5fd998fafb10c153a0070aa36c99ddfa869a2cbf2",
+        "scripts/rectification/decision_policy.py": "80e8de4680d19a052213bb821d2a9c94143a2962235aeb27f6ad81374d2e60d5",
+        "scripts/rectification/event_probes.py": "f5ca9d54501f67a1546fcf5b01da0c1277dec2e2c6df481884e8900d14909a7c",
+        "scripts/rectification/refinement_packet.py": "dbeea5aa780889785ce850f3881e8b7f67bb44594e9bc382c65b2e730884a2f4",
+        "scripts/rectification/scoring_service.py": "fa056fa30d9e6ffa4b75295fc49c793e35b26ebeca64ca843d26c8abef3050f2",
+        "scripts/research/cluster_width_lib.py": "1d8f2ffa7795d62069deebb2de8e16749317037d4ed3f24a2c0092676e5cfcc6",
+        "scripts/research/probe_supply_after_six.py": "098a5b4398d6f8d996917d809c2fc922ad64031d3b70660f8ef7c876aba92438",
+        "scripts/research/reported_offset_sweep.py": "1c7183e77f442b8c6b46761ff8a833b6c27b7841999426c72556237f45df1343",
+        "scripts/research/sealed_holdout_rerun.py": "957b824b7579c4f52681c85d7964fc802f8179470c93917cded63c7fe6c2769d",
+        "scripts/shadbala.py": "352a9dd91c9be6dd7a294b1dc24c62173e8f08db5fc36c453dc5dcd7e0fd62d7",
+        "scripts/varga.py": "1bca389e4f36d293a755de175163c3f39b42bab5831cb6f7828830899f9f177e"
+      },
+      "hash_scope": "explicit_identity_file_sets_not_a_transitive_dependency_lock"
+    },
+    "scorer_frozen_before_rerun": true,
+    "source_audit_status": "corrected_known_date_errors",
+    "trial_count": 20,
+    "official_valid_independent_blind": false,
+    "is_blind_evaluation": false,
+    "truth_hidden_from_ranker": true,
+    "results_previously_seen": true,
+    "must_not_claim_as_release_metrics": true,
+    "must_not_use_for_tuning": true,
+    "verified_minute_claim_allowed": false,
+    "metric_gates_passed": false,
+    "events_per_case": 3,
+    "boundary": "Previously exposed v3 shadow fixed-protocol rerun after the Moon Paksha convention fix (ERR-110, BUG-1557). Trials and metrics match the 2026-11-04 data_audit report on every field. Versus the 2026-09-20 historical report, changed_trial_count stays 8. Not a release metric. Prior reports and freezes preserved byte-for-byte.",
+    "report_sha256": "0dabbef823a65b089574d90bac6af3f903712684c8fd88b42cf424b650f3b532",
+    "freeze_record_sha256": "ad125812c8472048e012c8dbb3b7090bb3b94f2fff44ef8f3846a7d5a8f976c0",
+    "replay_started_at_utc": "2026-11-04T07:22:50.170060+00:00",
+    "replay_finished_at_utc": "2026-11-04T07:22:52.478378+00:00",
+    "previous_fixed_protocol_rerun": {
+      "report_path": "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json",
+      "freeze_record_path": "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.freeze.json",
+      "superseded_reason": "BUG-1557 (scripts/shadbala.py: the Moon's Paksha is no longer doubled; convention bphs_santhanam_27_10_11_chesta_27_18) changed a frozen identity file. Output is unchanged. Rectification scoring reads only Sthana, Drik and Naisargika, so ERR-110 stored a new record. Prior data_audit report and freeze preserved byte-for-byte."
+    }
+  },
+  "previous_fixed_protocol_rerun": {
     "evaluated_on": "2026-11-04",
     "report_path": "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json",
     "freeze_record_path": "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.freeze.json",
@@ -212,106 +312,6 @@
       "superseded_reason": "The data-audit fixes BUG-1545 (scripts/narayana_dasha.py, scripts/active_rectification_event_engine.py) and BUG-1546 (scripts/shadbala.py) changed frozen identity files. Output is unchanged, so ERR-110 stored a new record. Prior d60_mapping report and freeze preserved byte-for-byte."
     }
   },
-  "previous_fixed_protocol_rerun": {
-    "evaluated_on": "2026-10-29",
-    "report_path": "docs/research/sealed_holdout_rerun_d60_mapping_2026_10_29.json",
-    "freeze_record_path": "docs/research/sealed_holdout_rerun_d60_mapping_2026_10_29.freeze.json",
-    "implementation_sha256": "4bb731f7b5043de0b967ccd7e852ae25a1ecdce96c2b7987c142d7878b6632b2",
-    "extended_identity": {
-      "historical_artifacts_manifest_sha256": "6102a26a840be207b5858b3a4c0509274468ae9071e86d308cbbac7371d6864e",
-      "production_scoring_files": [
-        "scripts/active_rectification_event_engine.py",
-        "scripts/active_rectification_events.py",
-        "scripts/ashtakavarga.py",
-        "scripts/dasha_analyzer.py",
-        "scripts/divisional_charts_extended.py",
-        "scripts/domain_calculation_service.py",
-        "scripts/functional_benefics.py",
-        "scripts/jaimini.py",
-        "scripts/minute_rectification_fact_ranker_v4.py",
-        "scripts/minute_rectification_feature_facts_v4.py",
-        "scripts/narayana_dasha.py",
-        "scripts/rectification/api_service.py",
-        "scripts/rectification/candidate_contrast.py",
-        "scripts/rectification/candidate_window.py",
-        "scripts/rectification/case_holdout.py",
-        "scripts/rectification/contracts.py",
-        "scripts/rectification/dasha_transition_proximity.py",
-        "scripts/rectification/decision_policy.py",
-        "scripts/rectification/event_probes.py",
-        "scripts/rectification/refinement_packet.py",
-        "scripts/rectification/scoring_service.py",
-        "scripts/shadbala.py",
-        "scripts/varga.py"
-      ],
-      "production_scoring_sha256": "fcf10cb3a5742bb7e2d98d15d69638769cd830f470373ea93745a1cab3718c9b",
-      "research_files": [
-        "scripts/research/reported_offset_sweep.py",
-        "scripts/research/cluster_width_lib.py",
-        "scripts/research/probe_supply_after_six.py",
-        "scripts/research/sealed_holdout_rerun.py",
-        "scripts/minute_rectification_blind_eval.py",
-        "scripts/minute_rectification_fact_blind_eval_v4.py",
-        "scripts/minute_rectification_holdout_validator.py"
-      ],
-      "research_implementation_sha256": "4eb2c03b09a179dd81f912d735823dd09ec8506da2538f40d3a8e04d3ebc6c67",
-      "file_sha256": {
-        "scripts/active_rectification_event_engine.py": "96c537d221a84680ddaaba74e3335162c44b5bbd9a60d10e69550a2095910d98",
-        "scripts/active_rectification_events.py": "03e333ccf1103caf685dfe50fe890ab1793e945037085f50ce834a4a3b232222",
-        "scripts/ashtakavarga.py": "c524da3f54b2ed9c080b78b33854854a0ca3d5d79b7e536bb1c3bc2be3e02852",
-        "scripts/dasha_analyzer.py": "3949b96813217fe624899856afec9f0b787c4cb0fea3a9a0d961990dbd11943a",
-        "scripts/divisional_charts_extended.py": "e9820e80f14fddc86ab5495d11455d16c2d3c9502a9a242fcf9efc41186e0d11",
-        "scripts/domain_calculation_service.py": "66923a2bc69daf65c965d2973c4b9234edbec5147909eaf3609b6b377ce3c7ca",
-        "scripts/functional_benefics.py": "cbaac3b8c09b97f8724b5300e9a402cbf77089af6c7db8788550d40fd35ac57b",
-        "scripts/jaimini.py": "6b88658da5404da9b136686d92a04be462d61a39da6e8de4255021e2ed2d1173",
-        "scripts/minute_rectification_blind_eval.py": "07aa30d8bad3278669495d5fea8be937aaf6367c8af980ff4f2fe9aa84ccc6ff",
-        "scripts/minute_rectification_fact_blind_eval_v4.py": "bdf460e3467208e8b87345c5beb2128f3e22de7eeffd68f72b281b3ad9affc92",
-        "scripts/minute_rectification_fact_ranker_v4.py": "f1a0d8d2482cd754d2e7f253a9606b0373519435a2bf58cfc3753e2b9ae0fb76",
-        "scripts/minute_rectification_feature_facts_v4.py": "3d899a2cdd4825b8e32062443d487edcd2173a4ea55a36eb9f3abe587c0faf0f",
-        "scripts/minute_rectification_holdout_validator.py": "b60b1c752e58672beaee8b36eb3b2348169726cda070ec969db0f13e7ed715af",
-        "scripts/narayana_dasha.py": "bc667c30c9562fb73fb0b7b4c22a5e377f9a531795141fc629b2cd3cf396773a",
-        "scripts/rectification/api_service.py": "089730097d3872a4f7b5d7dd5574bb7e9ab4d2e246d4a380800bc824d7979b95",
-        "scripts/rectification/candidate_contrast.py": "eee7da71c0574a08876404b78ca111153fa95fb83ace82b9fdca141d1992e273",
-        "scripts/rectification/candidate_window.py": "2fbc5058b12049b15fca375cdc125b8283e19f9f3e4bedcc291e344501891bef",
-        "scripts/rectification/case_holdout.py": "fe0e699b617d9e216acdfeff87af4221c1d6f4c101bf33e1dbf14e2bcd75a8b5",
-        "scripts/rectification/contracts.py": "c6dc6bd230e38f9f2126c224aae9ddbd6415304c15340f8c7cd3f06f6ef3160f",
-        "scripts/rectification/dasha_transition_proximity.py": "5ce4ed27b0f06c681abc34e5fd998fafb10c153a0070aa36c99ddfa869a2cbf2",
-        "scripts/rectification/decision_policy.py": "80e8de4680d19a052213bb821d2a9c94143a2962235aeb27f6ad81374d2e60d5",
-        "scripts/rectification/event_probes.py": "f5ca9d54501f67a1546fcf5b01da0c1277dec2e2c6df481884e8900d14909a7c",
-        "scripts/rectification/refinement_packet.py": "dbeea5aa780889785ce850f3881e8b7f67bb44594e9bc382c65b2e730884a2f4",
-        "scripts/rectification/scoring_service.py": "fa056fa30d9e6ffa4b75295fc49c793e35b26ebeca64ca843d26c8abef3050f2",
-        "scripts/research/cluster_width_lib.py": "1d8f2ffa7795d62069deebb2de8e16749317037d4ed3f24a2c0092676e5cfcc6",
-        "scripts/research/probe_supply_after_six.py": "098a5b4398d6f8d996917d809c2fc922ad64031d3b70660f8ef7c876aba92438",
-        "scripts/research/reported_offset_sweep.py": "cabc96d1f8c7ed06ef34da4e02e052f598a5e62fd71a7ccbe7b0b099247c27d0",
-        "scripts/research/sealed_holdout_rerun.py": "69a7ddddfc4a88ec5c85f400c763f8cdbb3c58c00fe9e788a72f79daba1d3bc1",
-        "scripts/shadbala.py": "334dc9d8e4292f0c95d40fd4c29285aa89b9e49d4f187fb9498f4bb0ad93bdfc",
-        "scripts/varga.py": "1bca389e4f36d293a755de175163c3f39b42bab5831cb6f7828830899f9f177e"
-      },
-      "hash_scope": "explicit_identity_file_sets_not_a_transitive_dependency_lock"
-    },
-    "scorer_frozen_before_rerun": true,
-    "source_audit_status": "corrected_known_date_errors",
-    "trial_count": 20,
-    "official_valid_independent_blind": false,
-    "is_blind_evaluation": false,
-    "truth_hidden_from_ranker": true,
-    "results_previously_seen": true,
-    "must_not_claim_as_release_metrics": true,
-    "must_not_use_for_tuning": true,
-    "verified_minute_claim_allowed": false,
-    "metric_gates_passed": false,
-    "events_per_case": 3,
-    "boundary": "Previously exposed v3 shadow fixed-protocol rerun after the D60 mapping fix (ERR-110, BUG-1500). Trials and metrics match the 2026-10-18 upstream sync8 and the upstream_sync9b reports. Versus the 2026-09-20 historical report, changed_trial_count stays 8. Not a release metric. Prior reports and freezes preserved byte-for-byte.",
-    "report_sha256": "9e667c4e5819674e82519bfe9415d49a202515c765ea99609a9dd27feb69c77d",
-    "freeze_record_sha256": "8f300876eca9aac683879f1b26204be885c03b2c7b6d03e5930386852cbe4700",
-    "replay_started_at_utc": "2026-10-29T06:07:59.312582+00:00",
-    "replay_finished_at_utc": "2026-10-29T06:08:15.756678+00:00",
-    "previous_fixed_protocol_rerun": {
-      "report_path": "docs/research/sealed_holdout_rerun_upstream_sync9b_2026_10_28.json",
-      "freeze_record_path": "docs/research/sealed_holdout_rerun_upstream_sync9b_2026_10_28.freeze.json",
-      "superseded_reason": "The D60 mapping fix (BUG-1500) changed scripts/varga.py, which is part of the extended identity. Output is unchanged, so ERR-110 stored a new record. Prior upstream_sync9b report and freeze preserved byte-for-byte."
-    }
-  },
   "current_tree_unfrozen_diagnostic": {
     "is_blind_evaluation": false,
     "results_already_seen": true,
@@ -330,6 +330,28 @@
     "verified_minute_claim_allowed": false
   },
   "current_tree_reported_offset_replay": {
+    "report_path": "docs/research/reported_offset_moon_paksha_2026_11_04.json",
+    "report_sha256": "f0b602a0477d5913d4befcdf888f3687062c3e62e0b11e6ea3bc69303f0f02cb",
+    "freeze_record_path": "docs/research/reported_offset_moon_paksha_2026_11_04.freeze.json",
+    "freeze_record_sha256": "9728da404c736a85179dfbbd62ac551550ff8590494f4da62df874cac5ffbd49",
+    "scope": "reported_offset_sensitivity_not_product_accuracy",
+    "trial_count": 900,
+    "official_valid_independent_blind": false,
+    "official_blind_trial_count": 0,
+    "must_not_use_for_tuning": true,
+    "must_not_claim_as_release_metrics": true,
+    "replay_started_at_utc": "2026-11-04T07:22:52.718578+00:00",
+    "replay_finished_at_utc": "2026-11-04T07:25:28.356370+00:00",
+    "boundary": "Previously exposed reported-offset sensitivity after the Moon Paksha convention fix (ERR-110, BUG-1557). Summary and all 900 trials match the 2026-11-04 data_audit report on every field. Versus the 2026-09-20 historical report, changed_trial_count stays 581 of 900. Not a release metric. No tuning. Prior reports and freezes preserved byte-for-byte.",
+    "previous_reported_offset_replay": {
+      "report_path": "docs/research/reported_offset_data_audit_2026_11_04.json",
+      "report_sha256": "227498db6488da6f4841c14ea063abcd3b788a069f8e61bb3c384b8bcee55f94",
+      "freeze_record_path": "docs/research/reported_offset_data_audit_2026_11_04.freeze.json",
+      "freeze_record_sha256": "389995d59467aac0270c1395200126c9065340c8177362566bd816f8c4ef9321",
+      "superseded_reason": "BUG-1557 (scripts/shadbala.py: the Moon's Paksha is no longer doubled; convention bphs_santhanam_27_10_11_chesta_27_18) changed a frozen identity file. Summary and all 900 trials are unchanged. Rectification scoring reads only Sthana, Drik and Naisargika, so ERR-110 stored a new record. Prior data_audit report and freeze preserved byte-for-byte."
+    }
+  },
+  "previous_reported_offset_replay": {
     "report_path": "docs/research/reported_offset_data_audit_2026_11_04.json",
     "report_sha256": "227498db6488da6f4841c14ea063abcd3b788a069f8e61bb3c384b8bcee55f94",
     "freeze_record_path": "docs/research/reported_offset_data_audit_2026_11_04.freeze.json",
@@ -350,27 +372,5 @@
       "freeze_record_sha256": "20e7684bc6c7356a421c2c20517c153279561f810cd15d77f260c9915352aed9",
       "superseded_reason": "The data-audit fixes BUG-1545 (scripts/narayana_dasha.py, scripts/active_rectification_event_engine.py) and BUG-1546 (scripts/shadbala.py) changed frozen identity files. Summary is unchanged and 2 of 900 trials changed true_rank only (BUG-1545), so ERR-110 stored a new record. Prior d60_mapping report and freeze preserved byte-for-byte."
     }
-  },
-  "previous_reported_offset_replay": {
-    "report_path": "docs/research/reported_offset_d60_mapping_2026_10_29.json",
-    "report_sha256": "f40157139733552dbefdfc999f95458de75594cd511df0e5377d606e27559f0d",
-    "freeze_record_path": "docs/research/reported_offset_d60_mapping_2026_10_29.freeze.json",
-    "freeze_record_sha256": "20e7684bc6c7356a421c2c20517c153279561f810cd15d77f260c9915352aed9",
-    "scope": "reported_offset_sensitivity_not_product_accuracy",
-    "trial_count": 900,
-    "official_valid_independent_blind": false,
-    "official_blind_trial_count": 0,
-    "must_not_use_for_tuning": true,
-    "must_not_claim_as_release_metrics": true,
-    "replay_started_at_utc": "2026-10-29T06:08:17.701243+00:00",
-    "replay_finished_at_utc": "2026-10-29T06:29:23.437778+00:00",
-    "boundary": "Previously exposed reported-offset sensitivity after the D60 mapping fix (ERR-110, BUG-1500). Summary and trials match the 2026-10-18 upstream sync8 and the upstream_sync9b reports. Versus the 2026-09-20 historical report, changed_trial_count stays 581 of 900. Not a release metric. No tuning. Prior reports and freezes preserved byte-for-byte.",
-    "previous_reported_offset_replay": {
-      "report_path": "docs/research/reported_offset_upstream_sync9b_2026_10_28.json",
-      "report_sha256": "880bebc2e217100994a1442e47ead3f9d33e48aa3ac20f69d993f38def3ef25c",
-      "freeze_record_path": "docs/research/reported_offset_upstream_sync9b_2026_10_28.freeze.json",
-      "freeze_record_sha256": "385817bb27f04afeba230af5e547dd10dc8cf6db4b25c18bebe4f4e674cab972",
-      "superseded_reason": "The D60 mapping fix (BUG-1500) changed scripts/varga.py, which is part of the extended identity. Output is unchanged, so ERR-110 stored a new record. Prior upstream_sync9b report and freeze preserved byte-for-byte."
-    }
   }
 }

scripts/research/reported_offset_sweep.py

@@ -39,8 +39,11 @@ MINUTE_STEP = 1
 # 原值: docs/research/reported_offset_d60_mapping_2026_10_29.{freeze.json,json}
 # 新值: docs/research/reported_offset_data_audit_2026_11_04.{freeze.json,json}
 # 原因: 数据核对 BUG-1545（narayana_dasha.py、active_rectification_event_engine.py）与 BUG-1546（shadbala.py）改了冻结身份文件，ERR-110 另存新记录
-FREEZE = ROOT / "docs/research/reported_offset_data_audit_2026_11_04.freeze.json"
-REPORT = ROOT / "docs/research/reported_offset_data_audit_2026_11_04.json"
+# 原值: docs/research/reported_offset_data_audit_2026_11_04.{freeze.json,json}
+# 新值: docs/research/reported_offset_moon_paksha_2026_11_04.{freeze.json,json}
+# 原因: BUG-1557 月亮 Paksha 口径（scripts/shadbala.py）改了冻结身份文件，ERR-110 另存新记录
+FREEZE = ROOT / "docs/research/reported_offset_moon_paksha_2026_11_04.freeze.json"
+REPORT = ROOT / "docs/research/reported_offset_moon_paksha_2026_11_04.json"
 LEGACY_REPORT = ROOT / "docs/research/reported_offset_2026_09_20.json"
 
 

scripts/research/sealed_holdout_rerun.py

@@ -36,8 +36,11 @@ DATASET = ROOT / "references/real_case_calibration/minute_rectification_holdout_
 # 原值: docs/research/sealed_holdout_rerun_d60_mapping_2026_10_29.{freeze.json,json}
 # 新值: docs/research/sealed_holdout_rerun_data_audit_2026_11_04.{freeze.json,json}
 # 原因: 数据核对 BUG-1545（narayana_dasha.py、active_rectification_event_engine.py）与 BUG-1546（shadbala.py）改了冻结身份文件，ERR-110 另存新记录
-FREEZE = ROOT / "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.freeze.json"
-REPORT = ROOT / "docs/research/sealed_holdout_rerun_data_audit_2026_11_04.json"
+# 原值: docs/research/sealed_holdout_rerun_data_audit_2026_11_04.{freeze.json,json}
+# 新值: docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.{freeze.json,json}
+# 原因: BUG-1557 月亮 Paksha 口径（scripts/shadbala.py）改了冻结身份文件，ERR-110 另存新记录
+FREEZE = ROOT / "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.freeze.json"
+REPORT = ROOT / "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04.json"
 LEGACY_REPORT = ROOT / "docs/research/sealed_holdout_rerun_2026_09_20.json"
 ARCHIVE = ROOT / "docs/research/history/rectification_pre_cross_midnight_2026_09_20"
 PRODUCTION_FILES = [

scripts/shadbala.py

@@ -387,14 +387,25 @@ def _days_elapsed_since_base(year: int, base_year: int = 1951, base_days: int =
     return base_days + leap_years * 366 + (years - leap_years) * 365
 
 
+# BUG-1557: the Moon's Paksha convention of the precise path, named so it is
+# traceable. BPHS (Santhanam tr.) 27.10-11: benefics take the folded
+# elongation / 3, malefics 60 minus that, the Moon by phase (3.11) and not
+# doubled; 27.18: the Moon's Chesta Bala is her Paksha Bala. Her Paksha
+# therefore counts twice in the total (Kala + Chesta), as in saravali.github.io,
+# Maitreya8 (default) and jyotishyamitra. Raman's Graha and Bhava Balas doubles
+# it instead and leaves the Moon's Chesta out (Ex. 56); using both gave 4x.
+MOON_PAKSHA_CONVENTION = "bphs_santhanam_27_10_11_chesta_27_18"
+
+
 def paksha_phase(sun_longitude: float, moon_longitude: float) -> float:
     """Paksha base value: the Sun-Moon elongation folded to 0-180 degrees, / 3.
 
     BUG-1546: the elongation was taken as abs(Sun - Moon) without folding, so
     whenever the raw difference passed 180 degrees benefics got more than 60,
     malefics went negative and the doubled Moon value passed 120 (5 of 19
-    benchmark charts; Monroe's Moon -45.76). Raman, Graha and Bhava Balas:
-    if the difference exceeds 180 degrees, subtract it from 360.
+    benchmark charts; Monroe's Moon -45.76). Raman, Graha and Bhava Balas
+    and BPHS 27.10: if the difference exceeds 180 degrees (6 signs), subtract
+    it from 360 (12 signs).
     """
     elongation = abs(float(sun_longitude) - float(moon_longitude)) % 360.0
     if elongation > 180.0:
@@ -442,9 +453,12 @@ def calc_kala_bala_precise(pname: str, all_planets: Dict, context: Dict) -> Dict
 
     benefics, malefics = classify_drik_planets(all_planets)
     phase = paksha_phase(_planet_longitude("Sun", all_planets), _planet_longitude("Moon", all_planets))
+    # Waxing Moon is a benefic, waning Moon a malefic (BPHS 3.11, Santhanam:
+    # "decreasing Candr" is a malefic); classify_drik_planets already does this.
     paksha = phase if pname in benefics else round(60.0 - phase, 2) if pname in malefics else 0.0
-    if pname == "Moon":
-        paksha = round(paksha * 2.0, 2)
+    # 原值: 月亮 paksha ×2（Raman GBB §55(c)），Chesta 又取这个 ×2 后的值，总分里月亮盈亏力 = 4P
+    # 新值: 月亮 paksha 不乘 2；Chesta = Paksha（BPHS 27.18），总分里 = 2P
+    # 原因: BUG-1557，产品 2026-11-04 采用 MOON_PAKSHA_CONVENTION（BPHS-Santhanam 27.10–11 + 27.18）
 
     sunrise, sunset = context["sunrise_hour"], context["sunset_hour"]
     day_length = sunset - sunrise
@@ -516,6 +530,7 @@ def calc_chesta_bala_precise(pname: str, all_planets: Dict, kala: Dict, context:
     if pname == "Sun":
         return round(float(kala["ayana"]), 2)
     if pname == "Moon":
+        # BPHS 27.18: the Moon's Paksha Bala is her Chesta Bala (MOON_PAKSHA_CONVENTION).
         return round(float(kala["paksha"]), 2)
     sun_mean = _surya_mean_longitude("Sun", context)
     planet_mean = _surya_mean_longitude(pname, context)
@@ -640,6 +655,7 @@ def calc_shadbala(planets: Dict, asc_sign: str, birth_hour: float,
         'method_variants': {
             'kala': 'bphs_local_solar_events_ahargana_declination' if context else 'legacy_approximation',
             'chesta': 'bphs_bounded_surya_mean_motion_seeghrochcha' if context else 'legacy_speed_bands',
+            'moon_paksha': MOON_PAKSHA_CONVENTION if context else 'legacy_approximation',
         },
         'external_parity_boundary': 'Chesta remains cross-engine method-conflicted; totals inherit that boundary.',
         'is_night_birth': is_night,

scripts/vp_jain_shadbala_benchmark.py

@@ -49,6 +49,11 @@ def _value(row: dict, path: tuple[str, ...]) -> float:
 def _variant(component: str, planet: str) -> str:
     if component == "sthana":
         return "moolatrikona_degree_range_vs_whole_sign"
+    if component == "kala" and planet == "Moon":
+        # BUG-1557: the fixture doubles the Moon's Paksha (Raman GBB) and gives her
+        # Chesta 0; the engine follows BPHS-Santhanam 27.10-11 + 27.18 (P in Kala,
+        # P again as Chesta). The Moon's Paksha total is 2P on both sides.
+        return "moon_paksha_convention"
     if planet in {"Sun", "Moon"}:
         return "luminary_chesta_policy"
     return "mean_motion_seeghrochcha_variant"

tests/test_sealed_holdout_contract_freshness.py

@@ -68,6 +68,11 @@ IDENTITY_FORBIDDEN_PREFIXES = (
     # 原因: 数据核对 BUG-1545 / BUG-1546 改了冻结身份文件，ERR-110 另存新记录
     "docs/research/sealed_holdout_rerun_data_audit_2026_11_04",
     "docs/research/reported_offset_data_audit_2026_11_04",
+    # 原值: 排除到 data_audit_2026_11_04
+    # 新值: 再排除 moon_paksha_2026_11_04 两份冻结路径（追加，不替换上面的）
+    # 原因: BUG-1557 月亮 Paksha 口径改了冻结身份里的 shadbala.py，ERR-110 另存新记录
+    "docs/research/sealed_holdout_rerun_moon_paksha_2026_11_04",
+    "docs/research/reported_offset_moon_paksha_2026_11_04",
     "docs/research/history/",
     "artifacts/",
 )

tests/test_shadbala_moon_paksha_convention.py

@@ -0,0 +1,84 @@
+"""BUG-1557: the Moon's Paksha counts 2P in Shadbala, not 4P.
+
+The precise path doubled the Moon's Paksha in Kala (Raman, Graha and Bhava
+Balas §55(c)) and then took that doubled value again as her Chesta (BPHS
+27.18), so her Paksha entered the total four times. Product decision
+2026-11-04: BPHS (Santhanam) 27.10-11 + 27.18 -- not doubled, Chesta = Paksha,
+waxing Moon benefic, waning Moon malefic (3.11).
+"""
+
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+from types import SimpleNamespace
+
+import pytest
+
+ROOT = Path(__file__).resolve().parents[1]
+if str(ROOT / "scripts") not in sys.path:
+    sys.path.insert(0, str(ROOT / "scripts"))
+
+import jyotish_engine  # noqa: E402
+from shadbala import MOON_PAKSHA_CONVENTION, classify_drik_planets, paksha_phase  # noqa: E402
+
+# Public Rodden AA charts (Astro-Databank), Raman ayanamsa, mean node.
+CHARTS = {
+    # Steve Jobs, San Francisco: waxing, elongation about 32 degrees.
+    "jobs": dict(year=1955, month=2, day=24, hour=19, minute=15, lat=37.7749, lon=-122.4194, tz=-8.0),
+    # Marilyn Monroe, Los Angeles: waning (raw difference past 180 degrees).
+    "monroe": dict(year=1926, month=6, day=1, hour=9, minute=30, lat=34.0522, lon=-118.2437, tz=-8.0),
+    # Barack Obama, Honolulu: waning; the old 4x value gave Chesta 73.88 and Kashta 0.
+    "obama": dict(year=1961, month=8, day=4, hour=19, minute=24, lat=21.3069, lon=-157.8583, tz=-10.0),
+}
+
+
+def _chart(name: str) -> tuple[dict, dict]:
+    args = SimpleNamespace(second=0, ayanamsa="raman", node_mode="mean", **CHARTS[name])
+    chart, _, _, _ = jyotish_engine._compute_chart_from_args(args)
+    return chart["planets"], jyotish_engine.cmd_shadbala(args)
+
+
+@pytest.mark.parametrize("name", sorted(CHARTS))
+def test_moon_paksha_is_p_and_chesta_equals_paksha(name: str) -> None:
+    planets, result = _chart(name)
+    moon = result["planets"]["Moon"]
+    base = paksha_phase(planets["Sun"]["degree"], planets["Moon"]["degree"])
+    benefics, malefics = classify_drik_planets(planets)
+    waxing = (planets["Moon"]["degree"] - planets["Sun"]["degree"]) % 360.0 <= 180.0
+    # BPHS 3.11: waxing Moon benefic, waning Moon malefic.
+    assert ("Moon" in benefics) is waxing and ("Moon" in malefics) is not waxing
+    expected = base if waxing else round(60.0 - base, 2)
+    assert moon["kala_bala"]["paksha"] == pytest.approx(expected, abs=0.011)  # P, not 2P
+    assert moon["chesta_bala"] == moon["kala_bala"]["paksha"]  # BPHS 27.18
+    assert 0.0 <= moon["chesta_bala"] <= 60.0
+    assert 0.0 <= moon["ishta_phala"] <= 60.0
+    assert 0.0 <= moon["kashta_phala"] <= 60.0
+    assert moon["ishta_phala"] + moon["kashta_phala"] > 0.0
+    assert result["method_variants"]["moon_paksha"] == MOON_PAKSHA_CONVENTION == "bphs_santhanam_27_10_11_chesta_27_18"
+
+
+def test_jobs_moon_values_are_pinned() -> None:
+    _, result = _chart("jobs")
+    moon = result["planets"]["Moon"]
+    # Before BUG-1557: paksha 21.34, chesta 21.34, total 659.40 Virupa.
+    assert moon["kala_bala"]["paksha"] == 10.67
+    assert moon["chesta_bala"] == 10.67
+    assert moon["total_virupas"] == pytest.approx(638.06, abs=0.011)
+
+
+def test_obama_moon_kashta_is_no_longer_clamped() -> None:
+    _, result = _chart("obama")
+    moon = result["planets"]["Moon"]
+    # Before BUG-1557: chesta 73.88 (> 60), ishta 64.99 (> 60), kashta clamped to 0.
+    assert moon["chesta_bala"] == 36.94
+    assert moon["kashta_phala"] > 0.0
+
+
+def test_other_planets_keep_their_paksha() -> None:
+    planets, result = _chart("jobs")
+    base = paksha_phase(planets["Sun"]["degree"], planets["Moon"]["degree"])
+    benefics, malefics = classify_drik_planets(planets)
+    for planet in ("Sun", "Mars", "Mercury", "Jupiter", "Venus", "Saturn"):
+        expected = base if planet in benefics else round(60.0 - base, 2)
+        assert result["planets"][planet]["kala_bala"]["paksha"] == pytest.approx(expected, abs=0.011), planet

tests/test_shadbala_paksha_fold.py

@@ -26,8 +26,10 @@ def test_kala_paksha_stays_in_range_when_the_raw_difference_passes_180():
     planets = _planets(350.0, 10.0)
     for planet in ["Sun", "Moon", "Mars", "Mercury", "Jupiter", "Venus", "Saturn"]:
         paksha = calc_kala_bala_precise(planet, planets, context)["paksha"]
-        limit = 120.0 if planet == "Moon" else 60.0
-        assert 0.0 <= paksha <= limit, (planet, paksha)
+        # 原值: 月亮上限 120（Paksha ×2）
+        # 新值: 七星上限都是 60
+        # 原因: BUG-1557 月亮 Paksha 不再乘 2（BPHS-Santhanam 27.10–11）
+        assert 0.0 <= paksha <= 60.0, (planet, paksha)
     # Jupiter is a natural benefic: its paksha is exactly elongation / 3 (the
     # unfolded formula gave 340 / 3 = 113.33 here).
     assert calc_kala_bala_precise("Jupiter", planets, context)["paksha"] == round(20.0 / 3.0, 2)

tests/test_shadbala_precise_context.py

@@ -72,7 +72,10 @@ def test_precise_kala_preserves_archived_comparison_with_corrected_hora() -> Non
         # BUG-1275: Ayana reads tropical = sidereal + ayanamsa. The fixed sidereal
         # longitudes of _planets() were captured in the old mean-ayanamsa frame; with the
         # true ayanamsa the Moon's Ayana moves 26.09 -> 26.08 (Kala -0.01).
-        delta = 52.17 if planet == "Moon" else -60.0 if planet == "Venus" else 0.0
+        # 原值: 月亮 delta 52.17（月亮 Paksha ×2 = 21.33）
+        # 新值: 月亮 delta 41.51（Paksha 10.66，不乘 2；PyJHora 仍乘 2，差 10.66）
+        # 原因: BUG-1557 采用 BPHS-Santhanam 27.10–11 + 27.18（月亮 Paksha 不乘 2，Chesta = Paksha）
+        delta = 41.51 if planet == "Moon" else -60.0 if planet == "Venus" else 0.0
         assert abs(calc_kala_bala_precise(planet, planets, context)["total"] - expected - delta) < 0.001
 
 
@@ -83,6 +86,8 @@ def test_precise_chesta_uses_classical_bounded_variant() -> None:
     result = {planet: calc_chesta_bala_precise(planet, planets, kala[planet], context) for planet in planets}
     assert result["Sun"] == kala["Sun"]["ayana"]
     assert result["Moon"] == kala["Moon"]["paksha"]
+    # BUG-1557: the Moon's Paksha is P (waxing, 32 degrees / 3), not 2P, so Chesta is P too.
+    assert result["Moon"] == 10.66
     for planet in ("Mars", "Mercury", "Jupiter", "Venus", "Saturn"):
         assert 0.0 <= result[planet] <= 60.0
     vedastro = {"Mars": 21.7377, "Mercury": 47.5107, "Jupiter": 46.0549, "Venus": 30.8469, "Saturn": 39.4390}

tests/test_shadbala_service_api.py

@@ -95,9 +95,13 @@ def test_birth_data_request_returns_precise_shadbala(api_base) -> None:
     assert payload["success"] is True
     result = payload["result"]
     # The birth moment switches Kala and Chesta to the Swiss Ephemeris context.
+    # 原值: method_variants 只有 kala、chesta
+    # 新值: 加 moon_paksha（具名口径）
+    # 原因: BUG-1557 月亮 Paksha 口径绑定 BPHS-Santhanam 27.10–11 + 27.18
     assert result["method_variants"] == {
         "kala": "bphs_local_solar_events_ahargana_declination",
         "chesta": "bphs_bounded_surya_mean_motion_seeghrochcha",
+        "moon_paksha": "bphs_santhanam_27_10_11_chesta_27_18",
     }
     assert sorted(result["ranking"]) == sorted(SEVEN)
     for planet in SEVEN:

tests/test_vp_jain_shadbala_benchmark.py

@@ -20,14 +20,22 @@ def test_vp_jain_published_component_benchmark_is_fully_classified() -> None:
     # 原因: TASK-upstream-sync4-20261002 T3（上游 0afa2780）平运动连续计时、短弧中点后，
     #       Chesta 火星 19.25→21.08、木星 11.23→8.86、土星 6.78→4.34 进入 ±1 Virupa；
     #       金星 27.23→30.03 离开（公布值 28.18）；水星 18.08→27.14 仍差 1.62。
-    assert report["summary"]["within_tolerance_count"] == 36
-    assert report["summary"]["method_variant_count"] == 6
+    # 原值: within_tolerance 36 / method_variant 6
+    # 新值: within_tolerance 35 / method_variant 7
+    # 原因: BUG-1557 月亮 Paksha 不再乘 2（BPHS-Santhanam 27.10–11 + 27.18），
+    #       月亮 Kala 205.96→151.58（公布值 205.85，按 Raman ×2），离开容差；
+    #       月亮 Chesta 108.76→54.38（公布值 0）仍是 luminary_chesta_policy。
+    assert report["summary"]["within_tolerance_count"] == 35
+    assert report["summary"]["method_variant_count"] == 7
 
 
 def test_vp_jain_benchmark_keeps_known_method_variants_explicit() -> None:
     rows = build_report()["rows"]
     variants = {(row["component"], row["planet"]): row for row in rows if row["status"] != "within_tolerance"}
 
+    # 原值: 非匹配集里没有 ("kala", "Moon")
+    # 新值: 加 ("kala", "Moon")；月亮 Kala + Chesta = 151.58 + 54.38 = 205.96，仍与公布 205.85 + 0 差 0.11
+    # 原因: BUG-1557 月亮 Paksha 从 Kala ×2 改为 Kala P + Chesta P，总分里同为 2P
     assert set(variants) == {
         ("sthana", "Sun"),
         ("sthana", "Mercury"),
@@ -35,8 +43,11 @@ def test_vp_jain_benchmark_keeps_known_method_variants_explicit() -> None:
         ("chesta", "Moon"),
         ("chesta", "Mercury"),
         ("chesta", "Venus"),
+        ("kala", "Moon"),
     }
     # 原值: chesta 非匹配集 {Sun, Moon, Mars, Mercury, Jupiter, Saturn}
     # 新值: {Sun, Moon, Mercury, Venus}（原因同上：T3 Chesta 修正后火/木/土落入容差、金星跳出）
     assert variants[("sthana", "Sun")]["variant"] == "moolatrikona_degree_range_vs_whole_sign"
     assert variants[("chesta", "Moon")]["variant"] == "luminary_chesta_policy"
+    # BUG-1557: fixture Moon Kala carries the doubled Paksha; the engine counts P in Kala + P as Chesta.
+    assert variants[("kala", "Moon")]["variant"] == "moon_paksha_convention"