Commit Graph
4032 Commits
Author SHA1 Message Date
jesse-uxandGrok (agent loop) bf5186fe62 revert(#170): drop the info-deleting compression, keep the measurement (round 2 start)
按 FIX-token-card-1:
- references/product_runtime_contract.md 恢复原样
- 删除自拟任务书(以 Claude 的 TASK-token-card-20261109 为准)
- 三处被弱化的断言恢复原样
- consultation-evidence-card.ts 回到 staging:删信息换来的压缩全部作废,重新按
  「不删信息」做
- 新增测量脚本 consult-card-size-170.mts(真实工具链,state.evidenceCardChars)

基线 c032823f 实测(3 盘 × 10 领域):卡片 min 41,963 / 中位 46,017 / max 53,930 /
平均 46,587;modelVisibleChars 平均 91,559。块大小表见 PROGRESS。

Co-Authored-By: Grok (agent loop)
2026-11-09 12:48:28 +08:00
jesse-ux ebe0797b36 Merge remote-tracking branch 'origin/staging' into codex/token-card-compress-20261109
# Conflicts:
#	docs/BUG_HISTORY.md
2026-11-09 12:46:41 +08:00
ClaudeandClaude Opus 5.5 abcdc6d4d3 docs(task): #170 Claude brief + fix brief FIX-token-card-1 (review fail on e5c13936)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 12:29:41 +08:00
ClaudeandClaude Opus 5.5 afac4ea92f fix(accept): #179 module moves into scripts/ondemand_calc (scripts-root budget)
Independent Staging Quality Gate / validate-frontend (push) Successful in 6m51s
Independent Staging Quality Gate / validate-python (push) Successful in 8m26s
Independent Staging Quality Gate / publish (push) Successful in 3m1s
test_python_hygiene_budget forbids new root-level scripts/*.py; the brief named
scripts/muhurta_search_api.py. Same code at scripts/ondemand_calc/muhurta_search.py;
route, CLI import and tests follow.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 10:07:39 +08:00
ClaudeandClaude Opus 5.5 c84d143088 fix(accept): #173 natal Moon is never the transit Moon; unknown ayanamsa blocks
transit_date counted house_from_moon from the transit Moon when the caller gave
no natal Moon; the natal reference is now only what the caller supplied.
resolve_profile now refuses an unsupported ayanamsa name instead of passing it
to the engine.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 09:58:59 +08:00
ClaudeandClaude Opus 5.5 a3339c2fed fix(accept): #169 fixture freshness compares floats to 9 decimals
The fixtures were captured on Windows; on Linux the Moon longitude differs in
the 13th digit (41.48538690748626 vs 41.485386907488675), so the exact-equality
freshness test failed. Rounding both sides to 9 decimals still catches any real
engine change.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 09:58:59 +08:00
Claude 8213aadad6 merge: codex/ondemand-transit-date-20261109 into acceptance batch 1109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8

# Conflicts:
#	docs/BUG_HISTORY.md
#	scripts/jyotish_api_server.py
2026-11-09 09:58:55 +08:00
Claude 5967b394eb merge: codex/counterexamples-20261109 into acceptance batch 1109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8

# Conflicts:
#	docs/BUG_HISTORY.md
2026-11-09 09:58:54 +08:00
Claude 1860b74e84 merge: codex/muhurta-20261109 into acceptance batch 1109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8

# Conflicts:
#	docs/BUG_HISTORY.md
2026-11-09 09:58:54 +08:00
Claude 137b964d80 merge: codex/dasha-dates-engine-only-20261109 into acceptance batch 1109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8

# Conflicts:
#	docs/BUG_HISTORY.md
2026-11-09 09:58:54 +08:00
Claude c723f7333d merge: codex/reports-no-web-wording-20261109 into acceptance batch 1109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8

# Conflicts:
#	docs/BUG_HISTORY.md
2026-11-09 09:58:53 +08:00
jesse-uxandGrok (agent loop) 99273b8ec2 fix(muhurta): per-day windows, per-window verdicts, per-day offsets (#179 round 2, BUG-1822)
按 FIX-muhurta-1 改掉打回的四点:

- 候选日不再由「日出 + 6 小时」单点分数筛选:每天单独取当天报告,用
  _select_recommended_windows 取候选;单点分数只留在 legacy_reference(not_final)
- 每个时段在起止两端各调一次 check_activity_muhurta,写 verdict_at_start /
  verdict_at_end / window_verdict;两端不同写 mixed 并点名变化要素;每天写
  rejected_windows(不吉时段、判定不宜、被 time_window 裁掉)
- 跨夏令时:每天用自己的偏移取日出与当天报告;profile 改为 tz_policy=per_day
  + first_day_offset
- time_window 生效并校验格式;删掉死代码
- CLI 打印时区策略与每段判定

Tests: 20 条(+6),api_server_growth_contract 9 条;pytest -k muhurta 122 过
5 败(5 条为基线既有失败,已在基线 commit 复现)。

Co-Authored-By: Grok (agent loop)
2026-11-09 09:21:11 +08:00
ClaudeandClaude Opus 5.5 2726b96a1b docs(task): fix brief FIX-muhurta-1 for #179 (review fail on 28c719cb)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 09:09:35 +08:00
jesse-uxandGrok (agent loop) 8dcc6537af feat(ondemand): shared on-demand calculation skeleton + transit on a date (#173, BUG-1816)
产品 2026-11-09 拍板 C 方案:本单只做后端共用模块与信封,前端另开单。

- scripts/ondemand_calc/contract.py:信封 jyotish.ondemand_calc.v1;必填校验不猜
  默认值;节点/位置口径缺省用产品默认值 mean/apparent 并标 profile_source;
  输入指纹 sha256;失败只留缺口,不回显推断日期
- registry.py:TOOLS + ondemand_calc_api 分派;未知 tool 阻塞并点名
- transit_date.py:九曜黄经/星座/度数/星宿与 pada/逆行/日速度,按传入的本命
  参考算宫位,UTC 与当地时间都给;position_mode=mean 返回 blocked
- 薄路由 /api/ondemand_calc(_load_local_module,无新增类方法)

Tests: contract 8 条、transit 12 条、api_server_growth_contract 9 条全过。

Co-Authored-By: Grok (agent loop)
2026-11-09 09:09:16 +08:00
ClaudeandClaude Opus 5.5 53b0609846 docs(task): #173 decision — backend shared module only (option C), explicit product defaults, mean position blocked
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 09:03:41 +08:00
jesse-uxandGrok (agent loop) e5c139367d perf(consult): card keeps only the route's layers (#170, BUG-1813)
Base card gated by EVIDENCE_CARD_SPECS[domain].route: dasha systems keep Vimshottari/Narayana only, yoga detail only on yoga domains, annual brief only on the annual route; yoga row planets become names. Model-visible contract+card 46,664 -> 32,153 chars on average (-31.1%), max 54,532 -> 43,419 over 3 public charts x 12 domains. New size/route test; three existing assertions updated with old/new/reason notes.

Co-Authored-By: Grok (agent loop)
2026-11-09 08:59:44 +08:00
jesse-uxandGrok (agent loop) 28c719cbde feat(muhurta): location-exact on-demand muhurta search (BUG-1822)
原做法把参考时刻当 UT 12:00、不用地点、五要素只取「日出 + 6 小时」一个点。

- 新增 scripts/muhurta_search_api.py:必填 activity/日期/地点/tz 或 timezone_id/
  岁差/交点,缺一项报错;每天取实际日出日落,取不到整天 blocked;每个候选时段
  在起止两端各取一次五要素与 Hora,时段内变化标出变化时刻;现有单点评分放进
  legacy_reference 并标 not_final
- 薄路由 /api/muhurta_search(_load_local_module,无新增类方法)
- CLI 加 --lat --lon --tz --timezone-id --ayanamsa --node-mode;无地点时首行写明
  近似值、不给推荐时段,并去掉 int(hour_from_sunrise + 6) 冒充 UT
- 全局时区推断收紧为「lat/lon 都给了才推断」,否则择时不给地点会崩

Tests: tests/test_muhurta_search_api_20261109.py 14 条;
test_api_server_growth_contract 9 条通过;pytest -k muhurta 116 过 5 败,
5 条在基线 b453505d 上同样失败(无关)。

Co-Authored-By: Grok (agent loop)
2026-11-09 08:50:19 +08:00
jesse-uxandGrok (agent loop) 74a739b2dc feat(card): seven acceptance counterexamples end to end (BUG-1812)
- career layer gets the D1 longitudes, so the paksha has one source (Obama
  base vs career mismatch fixed)
- card carries movable_lagna_moon.reason, Narayana ad_status / ad_reason and
  the mahadasha cycle_index; running line says "sub-period undecided"
- engine keeps the MD when the AD is undecided; timingFollowing skips
  zero_duration rows explicitly
- timingKeys allowlist: adstatus / adreason / cycleindex / zeroduration
- fixtures + capture script, 16 engine tests, 7 card tests, PROGRESS, BUG-1812

Co-Authored-By: Grok (agent loop)
2026-11-09 08:38:56 +08:00
jesse-uxandGrok (agent loop) 152b298dd3 fix(chart): Chara dasha dates come from the engine, not the page (BUG-1811)
占星师 2026-11-08 第五轮裁定 5:所有大运日期必须由原生大运引擎生成,显示层
不得自己按年数比例重算。BUG-1752 只修了星盘页的 Vimshottari 轨;Chara 轨仍在
浏览器里按 365.25 天/年从出生日推大运、按年数比例切分运,与报告页的公历周年
口径最多差一天。

- 新增 scripts/chart_dasha/chara_dates.py:给 jaimini 的十二段与每段十二个
  分运写 start / end(本地日历日)与 start_ts / end_ts(UTC 秒);[start, end),
  末段终点钉在父段终点,0 年段保留 start == end,平年 2 月 29 日落 28 日。
- /api/dasha/chara 返回前写这些字段并带 date_basis。
- rashi_dashas_native._chara 改读同一模块(报告输出不变,已与旧版逐段对照)。
- 星盘页 charaTrack 只复制引擎日期;删掉 addYears;没有日期的行不显示。
- 文案 CHARA_ANTARDASHA_DATE_NOTE 改为「与大运同一套边界」。
- 两份真实引擎响应 fixture 的 chara 段用 _compute_chara_dasha 重抓。
- BUG_HISTORY BUG-1811;PROGRESS-dasha-dates-engine-only-20261109。

验证:pytest 新用例 6 passed,相关 63 passed;node --test 新用例 4 pass、
既有 105 pass;tsc exit 0;lint 0 error。

Co-Authored-By: Grok (agent loop)
2026-11-09 08:32:21 +08:00
jesse-uxandGrok (agent loop) 2e9ca2d73a fix(reports): drop web-verification wording from user-visible report text (BUG-1810)
产品负责人 2026-11-08:用户看到的报告文字不出现「联网 / 外部资料 / 外部核验 /
MEVG / 真实案例 / Real Case / Global Web」,后台记录保留。

- 个人报告写作提示词:依据与边界改为「没有用户经历可对照」,置信度封顶 B 不变
- 年报:技法执行情况跳过 MEVG / Real Case 两行;没有纳入的删掉对应条目
- 合盘:技法执行情况删掉「外部资料与真实案例核对:未进行」整行
- 产品导出审计章节:这两行不再渲染,证据边界同类一句删除,packet 原样保留
- 质量门:只要求 Functional Benefic/Malefic 渲染;新增后台检查,要求 packet
  仍带这两行且它们没有出现在 markdown 里

Tests: 新增 frontend/tests/report-no-web-wording-20261109.test.ts(4)与
tests/test_report_no_web_wording_20261109.py(3);更新年报、合盘、审计章节、
质量门既有断言(原值/新值/原因已注明)。

Co-Authored-By: Grok (agent loop)
2026-11-09 08:20:50 +08:00
ClaudeandClaude Opus 5.5 c032823f9e research(rectification): reseal round7_integration_2026_11_09 on staging b453505d (BUG-1809)
Independent Staging Quality Gate / validate-frontend (push) Successful in 5m18s
Independent Staging Quality Gate / validate-python (push) Successful in 6m37s
Independent Staging Quality Gate / publish (push) Successful in 2m27s
Product owner approved 2026-11-09 ("agree new result"). Same protocol as
BUG-1720 / BUG-1759: c1a69445^ replay tree + float_split + round5 overlays +
b453505d scripts/references + a small replay overlay (new tag paths, retired
scorer zero-row adapter, RuntimeError guards). Product default strength scope
seed_only_v0, comparability policy common_evidence_coverage_v1.

- sealed 20: top-1 0.35 -> 0.30, top-3 0.45 -> 0.45, mean error 6.80 -> 6.75
- offset 900: top-1 0.0156 -> 0.0100, coverage 0.900 -> 0.883 (164/900 vs
  round5 tag; 57/900 over 4 cases vs pre-integration staging)
- live v5 231: best = truth 8 -> 8; 124 changes vs round5 tag, 115 vs
  pre-integration staging
- leaders: sealed 40/40 comparable; offset 863 comparable / 35 provisional /
  2 no single leader; live 226 / 5

Trial-for-trial identical to the integration agent's seed_only runs.
Sensitivity only, not accuracy (event scoring near chance, BUG-1644).
Older tags unchanged.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 07:28:04 +08:00
ClaudeandClaude Opus 5.5 7cdf83106a docs: task brief for #179 muhurta (2026-11-09)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 07:18:34 +08:00
ClaudeandClaude Opus 5.5 f5356c62c1 docs: task brief for #173 ondemand-transit-date (2026-11-09)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 07:18:29 +08:00
ClaudeandClaude Opus 5.5 41862634f3 docs: task brief for #169 counterexamples (2026-11-09)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 07:18:12 +08:00
ClaudeandClaude Opus 5.5 b453505d50 test: eslint-clean self-reference pipeline test
Independent Staging Quality Gate / validate-frontend (push) Successful in 5m16s
Independent Staging Quality Gate / validate-python (push) Successful in 6m41s
Independent Staging Quality Gate / publish (push) Successful in 2m49s
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 00:39:04 +08:00
ClaudeandClaude Opus 5.5 ae714e0349 fix(narayana): default strength scope back to seed_only_v0 pending the astrologer (BUG-1784)
Owner 2026-11-09: the R2 widening is on hold until the astrologer rules on it.
STRENGTH_SCOPE_DEFAULT returns to seed_only_v0. All R2 code stays: local
propagation, the selected / equal / unresolved statuses, and
seed_ad_start_p43d_v1 as a named non-default profile. profile_config and the
cache keys record strength_scope = seed_only_v0. R1 and R3 are unchanged.
Tests that pinned the widened scope now pass it explicitly, with
原值 / 新值 / 原因 notes. Evidence-card golden recaptured (TTL=0; only the
profile labels changed).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 00:21:12 +08:00
ClaudeandClaude Opus 5.5 8feabcc588 docs(task): brief for dasha-dates-engine-only-20261109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 00:14:47 +08:00
ClaudeandClaude Opus 5.5 9c7b35ef63 docs(task): brief for reports-no-web-wording-20261109
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 00:14:42 +08:00
ClaudeandClaude Opus 5.5 b7ba7f4e4f fix(consult): R3 witness fields and Narayana profile_config reach the real card (BUG-1786, BUG-1783)
Found in the round-7 cross-branch check: on public charts (Zidane, Obama)
the Neecha Bhanga row never reached yogas_detail. The compact detector entry
dropped formation_evidence, and the row sat past the 30-row cap. Keep the
witness fields when compacting and never cut a row that carries them (the
cap stays 30). The card's Narayana method_status also dropped the engine's
profile_config (timing key allowlist); carry it through.

Evidence-card golden recaptured with TTL=0 (Obama: row 30 is now Neecha
Bhanga); the load-dependent VedAstro closure reason is kept as before.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-09 00:03:21 +08:00
ClaudeandClaude Opus 5.5 b1a83761b3 feat(consult): 解读依据 rows read the receipt's result_line / display_status (BUG-1793)
The 简明解读 list's grey line is the receipt row's result_line (fallback: what
the layer computed); firm → 已算, conditional → 有条件. One backstage filter
(isBackstageAuditRow: external / MEVG / real-case, plus provider, engine,
packet, calibration, Western) is shared by withTechniqueEvidence and the list.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 23:46:30 +08:00
ClaudeandClaude Opus 5.5 a9111fae20 merge: codex/chat-mode-ui-1108 into round7 integration (BUG-1795..1798)
Contract keeps BUG-1794 (no web / external / MEVG / real-case wording in either
mode) with BUG-1795's 深度解读 / 简明解读 names; the missing-audit-table repair
prompt keeps BUG-1794's "no verification rows" wording under the new name.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 23:43:49 +08:00
ClaudeandClaude Opus 5.5 8aa0d0cbcd merge: codex/round7-rect-kp-1108 into round7 integration (BUG-1801..1804)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 23:41:07 +08:00
ClaudeandClaude Opus 5.5 94d5bd8008 merge: codex/round7-evidence-1108 into round7 integration (BUG-1789..1794)
Evidence card independence keeps both the engine's R3 `self_reference_only`
name list and the evidence branch's `self_reference` notes map.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 23:40:37 +08:00
ClaudeandClaude Opus 5.5 5b0c206ee8 docs: BUG-1789..1794 history with r8 blind-eval results
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 23:31:06 +08:00
ClaudeandClaude Opus 5.5 2fd91f2b9e fix(consult): fact guard skips a layer named only as the comparison (BUG-1789)
r8 blind eval: 「这一层给出的结构与 D1 同向:土星在巨蟹」 in a D10 line was read
as a D1 claim (3 false mismatches). A layer token followed by 同向 / 一致 /
对照 … no longer opens its own span; the matched-claim note is only shown on
mismatches.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:57:41 +08:00
ClaudeandClaude Opus 5.5 fc319ab5ec docs: BUG-1801..1804 rectification comparability, API guard, lookup budget, KP note
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:41:31 +08:00
ClaudeandClaude Opus 5.5 43d99de64b fix(rectification-api): never raise the engine's can_apply=false (BUG-1802)
The high-rigor path overwrote can_apply with the technique contract's
confirmation_allowed, so a passing external validation could turn a
low-confidence or provisional engine result into an applicable one.
Both must now agree; otherwise the contract continues rectification
with local_engine_cannot_apply.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:41:31 +08:00
ClaudeandClaude Opus 5.5 0df8a3283f fix(rectification): provisional vs comparable leader under unequal coverage (BUG-1801)
Raw totals are only compared when every candidate computed the same
(event, layer) evidence units. Otherwise winning_segment stays the raw
leader (provisional_leader) and comparable_leader re-ranks over the
units all candidates share; a missing unit is dropped for everyone,
never scored 0. The live stage Python engine now skips an event whose
Narayana is unavailable instead of scoring it 0, repeated facts count
once, and the stage labels an unequal-coverage leader as provisional.

占星师 2026-11-08 验收清单 §6 / §9 / §10.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:41:31 +08:00
ClaudeandClaude Opus 5.5 f60e49c59f docs: BUG-1785 rectification sensitivity of round-6 R1 + R2, grouped by case
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:39:01 +08:00
ClaudeandClaude Opus 5.5 d784e99867 feat(narayana): round-6 R1 default AD direction, R2 strength scope and local propagation, R3 self-reference
- R1: default AD direction rath_p50_md_parity_v1 (start parity kept as a research
  profile); full profile config + algorithm version on every Narayana output, cache
  keys (consult, api chart, probes, long report, synastry) and the rectification run.
- R2: one strength profile at the seed, AD start and p.43 (d); selected / equal /
  unresolved kept apart; unresolved nodes block only what depends on them (seed
  candidates kept, dated MDs before an unresolved length kept, AD-only loss in
  rectification keeps the MD).
- R3: Neecha Bhanga witness fields (reference, relative_house, self_reference,
  independent_confirmation_allowed, self_reference_only) from the main detector to
  the evidence card yogas_detail rows; self-reference-only never counts as support.
- Sync her b54812c3 KP clock validation; port her round-5 / round-6 tests.
- BUG-1783..1787; golden recaptured with TTL 0.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:38:00 +08:00
ClaudeandClaude Opus 5.5 d3a752f415 feat(consult): no web / external / MEVG / real-case wording in answers, either mode (BUG-1794)
产品 2026-11-08 决定,偏离占星师「缺口不能隐藏」裁定(第五轮裁定 6 / 8):产品
选择不向用户展示;后台保留记录. The plain-mode gap sentence (BUG-1757) and the
professional audit table's four verification rows are no longer asked for;
the receipt's MEVG / Real Case Calibration rows are unchanged, and 「不写已验证 /
已校准」 stays.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:28:31 +08:00
ClaudeandClaude Opus 5.5 1839ce918b feat(consult): receipt audit rows carry this chart's result_line and firm / conditional status (BUG-1793)
产品 2026-11-08(简明解读「解读依据(N 项)」): rows the evidence card speaks for
get techniqueAuditTable[i].result_line (one plain line built from card fields
only, never by the model) and display_status (firm / conditional from the
two-axis interpretation_status; Narayana with a method dispute -> conditional).
Provider, MEVG, calibration and Western rows get neither.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:28:31 +08:00
ClaudeandClaude Opus 5.5 df2d6fd58c fix(consult): contract note prefixes and no literal maxSteps on the fact repair (BUG-1789..1792)
Product-rule notes use the announced 「产品规则,不是占星师的方法:A-09..A-11」
prefix; the repair stream keeps the route's step settings (no literal maxSteps).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:19:14 +08:00
ClaudeandClaude Opus 5.5 063c9d1be7 research(consult): evidence traceability metric (BUG-1793)
占星师 2026-11-08 第六轮 §5: share of evidence-requiring judgments that trace
to the right layer, entity and reference, replacing raw counts. Same reader
as the runtime guard; reads the card the model saw from the payload.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:13:34 +08:00
ClaudeandClaude Opus 5.5 d72277a597 feat(consult): reject and regenerate answer sentences the evidence card contradicts (BUG-1789)
占星师 2026-11-08 第六轮 R4 / V6: a sentence whose explicit layer / entity
claim (varga sign, lagna, house lord, D1 house with 本命, Karakamsha sign /
house / lord / lord house, Arudha sign / house / lord) does not match the card
is rejected after the answer finishes, regenerated once, and dropped if the
rewrite still mismatches, adds an execution status or breaks the table. The
corrected answer replaces the streamed one; only validation steps are recorded.
Anonymous minimal counterexamples for the three r6 real errors fail on the
previous code and pass now. Harness records the guard and the pre-guard text.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:13:34 +08:00
ClaudeandClaude Opus 5.5 7dbf383cf1 feat(consult): evidence fact fields, derived deps, Karakamsha lord house, self-reference (BUG-1790..1792)
占星师 2026-11-08 第六轮 R3 / R4:
- independence gains derived (derived fact -> dependency); the Karakamsha
  chain (AK D9 sign -> D1 projection) is derived, its lord's placement a
  separate raw fact; evidenceCardFactRecords expands every item to the ten
  fields, raw -> relation / derived -> conclusion.
- base.natal.jaimini.karakamsha carries karakamsha_lord_natal_house and a
  reading line (113 of 210 blind answers copied the sign's house to the lord).
- yoga self-reference fields (self_reference / independent_confirmation_allowed
  / self_reference_only) read defensively: no support vote, her wording.
- runtime contract: plain layer names, mismatch rejection, self-reference rule.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:13:34 +08:00
ClaudeandClaude Opus 5.5 d9306658fb docs: BUG-1798 capability-audit note
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:10:23 +08:00
ClaudeandClaude Opus 5.5 3b8a9803f4 fix(consult): keep the capability audit clean of verification-gate tokens (BUG-1798)
The backend capability audit lists "Case Validation" as user-visible when
frontend source contains the gate's acronym; the row filter, comments and
the dev preview fixture now avoid the literal token.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:10:16 +08:00
ClaudeandClaude Opus 5.5 9795033da7 feat(consult): 解读方式 pill, 解读依据 list, no 联网 rows (BUG-1795..1798)
Owner-approved chat answer-mode UI (2026-11-08):
- Modes named 深度解读 (professional, default) / 简明解读 (plain) in the UI,
  the model-facing mode line and the runtime contract (names only).
- 解读方式 pill left of send (常用问题 pill look); menu reuses the preset
  popup rows; 功能吉凶口径 (PVR 表 30 / BPHS 例外表) under 深度解读 only;
  Drawer on phone. Remembered per user in localStorage; applies from the
  next question; 重新生成 uses the current choice.
- Mode resolution: this question's words -> pill -> 深度解读; history no
  longer sets the mode. Request carries answerMode (+ functionalProfile
  only under 深度解读); engine gets analysis_mode / functional_profile.
- Receipt records answerMode / functionalProfile; 简明解读 answers show a
  collapsed 解读依据 hairline list built from the receipt audit rows.
- External verification rows (联网 / MEVG / 真实案例) are dropped from the
  rendered 深度解读 audit table and never listed.
- Composer stacks when the single-row field would be under 64px (320px).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 22:07:21 +08:00
ClaudeandClaude Opus 5.5 04b4960933 docs: BUG-1773 / BUG-1774 history with r7 blind-eval results
Independent Staging Quality Gate / validate-frontend (push) Successful in 5m9s
Independent Staging Quality Gate / validate-python (push) Successful in 6m47s
Independent Staging Quality Gate / publish (push) Successful in 2m51s
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
2026-11-08 21:13:44 +08:00