Round-4 ruling Q4. scripts/narayana_rath.py builds the Rath sequence from the
rule ledger and book rows ported verbatim from upstream yinduzhanxing c7117f92
(sanjay_rath_narayana_book_ledger JSON + md, chara_narayana_book_golden_cases):
Table 10 order (Saturn -> Table 11, Ketu -> reversed), years by
jaimini_odd_footed_dignity_v2 with jaimini_dual_lord_source_v1, ADs twelve
equal parts from the MD sign (Table 13 / 11 / 14), start sign by the existing
select_narayana_seed_sign, ties reported as parameter_sensitive with both
sequences.
Given the book's start sign, Tables 15 / 17 / 18 match 36/36 MD signs and
years. The start sign does not: Table 17 picks Virgo (book Pisces), Table 18
ties (book Cancer). Task red line 2: the profile stays out of every
user-visible surface and out of scoring; T5 trial not run. A test pins that no
script imports it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Round-4 astrologer ruling Q3. tajika.calc_panchadhikari_year_lord and its
missing pieces (Tri-Rasi tables, planet index maps, ayanamsa-aware PyJHora
lookups) plus references/tajika_panchavargiya_rules.json (blob e8c15ac3) are
ported verbatim from yinduzhanxing c7117f92, with one keyword added
(muntha_sign_idx): upstream builds the Muntha-lord candidate from the annual
Lagna + years; ours stays natal Lagna + completed years (BUG-1212).
Both annual producers (solar_return_full_report, cmd_tajika) call the new
tajika_year_lord.select_annual_year_lord, so the pack conflict gate compares
the same rule. Output carries candidates / candidate_list with Panchavargiya
score and engine, aspect flags, selection_basis and a one-sentence basis
(zh/en). The product path always uses the native proxy (PyJHora is AGPL, absent
from the API image and CI); JYOTISH_YEAR_LORD_PYJHORA_REFERENCE=1 restores the
upstream PyJHora lookup for research. Card varshesha_basis and the report
'Selection basis' rows show the sentence; the BUG-1212 stand-in label is gone.
Mudda Dasha and Tri-Pataka follow the selected year lord.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
scripts/jaimini.py (BPHS 8-karaka order), scripts/ashtakavarga.py (calc_sodhita_av
removed) and scripts/shadbala.py (local house score renamed and dropped from the
result) are in the frozen production identity. Records re-frozen under the label
astrologer_rulings_batch1_2026_10_03 (--freeze, then replay; PYTHONHASHSEED=0,
chart cache TTL 0); the upstream_sync5 records become 'previous', byte-for-byte
preserved. Replays identical to upstream_sync5: reported offset 0 of 900 trials
changed, sealed holdout 20 trials and metrics identical. v5 77-case evaluation
unchanged (see PROGRESS). ALGORITHM_VERSION stays rectification-v5-matrix-scoring-10.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Astrologer ruling 甲5. scripts/pancha_mahapurusha.py is the only detector: kendras from
the Lagna; retrograde no longer cancels; combustion (shared orb table), malefic
conjunction and malefic graha drishti (Sun 7, Mars 4/7/8, Saturn 3/7/10) mark the
yoga reduced with reasons; D9 debilitation still cancels when navamsa_sign is given
(current behaviour, ruling silent, asked next round). _detect_yogas now passes the
Sun's longitude (combustion was never checked). mahapurusha_*, *_own, moolatrikona_*,
hamsa_trikona (Jupiter 1/5/9, not classical) and bvr_maalavya_yoga are reference_only;
yogas_doshas.calc_pancha_mahapurusha_yoga is kept as reference_yogas only. The rule
engine appends main-detector rows; cmd_yoga rows now carry lon + retrograde so the
report and card get the same combustion verdict. Card label lists 部分减弱 reasons.
Changed assertions (three columns in PROGRESS): self-conjunction tests (malefic in
same sign: invalid -> reduced), run_all t31/t32 (combust/retro: invalid -> reduced /
full), run_all t54 (evaluated rules >= 400 -> evaluated + reference_only >= 400).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Astrologer ruling 甲7: yoga_expansion.detect_kemadruma is the main detector (2nd/12th
from the Moon count only Mars..Saturn, no kendra cancellation, nodes in kendras do
not cancel). kemadruma_yoga (ky2 kendra cancellation) and kemadruma_variant (nodes as
neighbours) are reference_only; YogaEngine.detect appends the main row.
BUG-1187 omission: the algorithmic _detect_solar_lunar_yogas block still counted the
Sun, Moon and nodes, so Ketu next to the Moon gave 'Anapha Yoga' beside Kemadruma
(Taylor) and Sun/nodes gave Durudhura beside Sunaphaa. Both blocks now count the
five true planets. divisional_yoga compared a 'house' key vargas never carry (always
'isolated') and counted the Sun; it now uses sign adjacency and the same five.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Astrologer ruling 乙14. One condition source, yogas_doshas.calc_nicha_bhanga_raj_yoga
(conditions A-F, now also structured). New scripts/main_yoga_detectors.py turns it
into yoga_engine-shaped rows (zh + en twins) that YogaEngine.detect appends; the
four neechabhanga_* rules are reference_only (kept, renamed, not evaluated).
Card: traditional_yogas canonical 落陷取消（Neecha Bhanga） with 成立条件; compact
yoga entries keep conditions/reductions. Report: category 落陷取消, mixed group,
not raja; personal-report alias neecha_bhanga -> special.
Changed assertion (test_consultation_yoga_detection_merge): canonical name
'Neecha Bhanga Raja Yoga' -> '落陷取消（Neecha Bhanga）' (ruling 乙14).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
scripts/jaimini.py (Chara Karaka 8 Rahu reversal), scripts/ashtakavarga.py
(Trikona / Ekadhipatya Shodhana, unwired) and scripts/rectification/
decision_policy.py (R1 floor, BUG-1193) are in the frozen production identity.
Records re-frozen under the label upstream_sync5_2026_10_02 (--freeze, then
replay; PYTHONHASHSEED=0, chart cache TTL 0); the upstream_sync4 records become
"previous", byte-for-byte preserved. Replays identical to upstream_sync4:
reported offset 0 of 900 trials changed, sealed holdout 20 trials identical,
metrics equal. ALGORITHM_VERSION stays rectification-v5-matrix-scoring-10.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Preta Yoga hard-coded the 11th house as Badhaka; it now uses the single
yoga_engine.badhaka_house() definition (movable 11 / fixed 9 / dual 7).
Remaining early-death / lineage-ending phrases removed from curse yoga
family effects and the Matru-hani yoga effects list.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
scripts/shadbala.py is in the frozen scoring identity (dataset
frozen_scoring.files), so the ported Kala/Chesta fixes and the BPHS Sun
minimum (BUG-1188/1189) turned the integrity gate red although the scorer only
reads the Sthana/Drik/Naisargika components. Records re-frozen under the label
upstream_sync4_2026_10_02 (--freeze, then replay; PYTHONHASHSEED=0, chart cache
TTL 0); the functional_v2 records become "previous", byte-for-byte preserved.
Replays are identical to functional_v2: reported offset 0 of 900 trials
changed, sealed holdout 20 trials identical; memoization golden v2 scores
unchanged and the v5 77-case evaluation is byte-identical to a18f8d4e, so
ALGORITHM_VERSION stays rectification-v5-matrix-scoring-10 (decision 6 only
bumps on a score change).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Follow-up to f3ed9509 (TASK-upstream-sync4-20261002 T5). The port dropped
combo_template_en from bvr_002/003/004/016/017/018, so the English edition had
no combination text for them; restored. kemadruma_yoga's text still said "only
the Sun and Moon" although the nodes no longer count. New tests: Ketu alone
next to the Moon leaves Kemadruma in place in both the rule and
detect_kemadruma; Saturn there cancels it and forms Sunaphaa.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
bvr_002..004 (Sunaphaa/Anaphaa/Duradhara) and bvr_016..018 (Vesi/Vosi/
Ubhayachara) take upstream yinduzhanxing 2dd6062a's logic (Mars..Saturn
only). kemadruma_yoga also ignores the nodes (product decision: classical
definition; upstream still counts them). The six lunar/solar duplicates of
upstream's RETIRED_RULE_IDS are retired; upstream's four other retirements
are out of scope. TASK-upstream-sync4-20261002 T5.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Functional roles feed the *_functional_*_auxiliary rules, so 57782aea changes
candidate scores for identical input (memoization fixture 12:00: 8.6274 ->
8.4977). Per the "scoring semantics change => bump ALGORITHM_VERSION"
precedent (scoring-7 -> 8 -> 9), the identity moves to scoring-10; policy v3,
input contract v5 and Skill versions are unchanged, history is not relabeled.
Five frontend sites and one SQL guard tested `=== "...scoring-9"` for the
dated candidate-window contract; they now use isDatedScoringAlgorithmVersion /
a generation regex (>= 9). Migration 20261002010000 only recreates
validate_dated_rectification_candidate (one-line guard change).
Memoization golden v2 written by the test's own write_golden; v1 (scoring-8)
frozen by sha256. Real-engine scoring-10 cross-midnight golden added. Research
records re-frozen per ERR-110 (label functional_v2_2026_10_02) and
scripts/functional_benefics.py added to the frozen production identity
(ERR-114: 57782aea changed scores without tripping it).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Ported verbatim from the upstream skill repo (yinduzhanxing e9beae6b) with
product authorization 2026-10-02: scripts/functional_benefics.py,
scripts/pancha_mahapurusha.py, references/functional-role-profile.md; only
the two functional-role methods of scripts/yoga_engine.py (our file diverged).
v5 77-case rectification eval: 5 cells up (to +7.8pp), one cell -1.3pp (one
case); product accepted. Rectification memoization golden still to be
regenerated (scores change with the grouping).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
yoga_rules.json gains effects_en for every rule and an _en sibling for each
Han-bearing string (combo_template, strength, source, accuracy_note); zh values
untouched. reader_dasha_applicability renders language='en' from the same
packet; the default zh output is byte-identical (pinned by sha256).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Wording only: _quality_user_meaning names start/change/interruption as
升学/学业变动/学业中断; _display_date_label drops the stored month for
year-precision events. Split hash and month field unchanged; same-machine
A/B (PYTHONHASHSEED=0) differs only in user_meaning/display_date_label.
Re-frozen per ERR-110 under new quality_wording_2026_09_29 records;
sealed rerun (20) and reported-offset sweep (900) identical to 09-21.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
The natal system prompt now says: the card in claim_cards is the chart
evidence for this answer, quote it as given, look up one further section of
this calculation with read-consultation-evidence before writing, and read
evidence_card.backstage as a confidence cap. The "use every executed layer"
and must_use_layers sentences are replaced; local_layers paths now point at
the card. SKILL.md 关联技法完整调取 / 0.0.1 and router 0.7 state that the
full result stays in receipts, the 本轮技法 panel and reports while web chat
answers from the card; computing the full spectrum is unchanged. Skill and
package version 6.9.16 -> 6.9.17 (tests/run_all.py 三栏).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
Carry explicit local date intervals instead of inferring the day from clock
order. Cluster width, delivery, adoption, and reports keep the actual civil
date; adopted date is stored separately from the reported birth_date.
Algorithm identity is scoring-9 / spec-v5. Scoring weights, confirmation
thresholds, and Skill version are unchanged. Isolated Linux final-3 gates
passed; four pre-existing Python failures remain. This is not a production
release.
Add date-isolated caches and regression coverage, align scoring identity, and freeze full research reruns while preserving historical artifacts. Record unresolved cache/receipt identity and end-to-end acceptance gaps for branch review only.
Co-Authored-By: Claude Code <noreply@anthropic.com>
Build holdout v4 from the public AA set, correct the two v3 dates, and sweep
R1–R5 plus pairs offline. No production scoring defaults change. No
implementation brief: delivered width stays the full window on every radius.
Venus/Jupiter/Moon periods no longer collapse into one marriage-opportunity line. Full reading reports the ten BPHS conditional dasha families; Skill 6.9.16.
Co-authored-by: Cursor <cursoragent@cursor.com>
Three-way-merge calculation modules and pl9-export into the product fork while keeping commercial API routes, Raman ayanamsa, and the consultation contract as a keypath superset.
Co-authored-by: Cursor <cursoragent@cursor.com>
Advance the one-way import to git commit a6f47abd with consultation keypath golden, switch official VedAstro comparison to the REST Calculate bridge, and receive the 25 new registry entries behind research_only_blocked.
Co-authored-by: Cursor <cursoragent@cursor.com>
One "一时说不好" no longer ends the interview. Plateau streaks use the
same sample floor. Delivery copy states why the range stopped.
Co-authored-by: Cursor <cursoragent@cursor.com>
Two real local sessions are archived. In them the agent states twice that
its minute came from a report already in the upstream repo rather than
from the user's answers, and the follow-up "fix" wired that answer into
the scoring chain; upstream now carries a Narayana tie-break derived from
that single case. This repo was checked and is clean, so two red lines now
keep it that way and gate any upstream sync.
The same records show what the answers genuinely bought: a 30-minute
window narrowed to 3. Tasks 7-10 add what makes that deliverable — the
per-answer narrowing table, an explain surface, mid-case window changes
with evidence retained, and the batched first round.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0155nFCgCHtoA7jhSDGmZmMu
The upstream interview playbook and evidence thresholds resolve task 0.
Upstream stops and reports when evidence is thin — fewer than three dated
events, fewer than two domains, or a tie — while this repo treats the same
kind of thresholds as a confirmation gate and keeps asking when they are
not met. Its label ladder has no confirmed rung at all, and every contract
test asserts candidate_range_not_birth_time_truth.
Task 1 capped the interview by round count, which is half of it. Task 6
adds the evidence-state stop rules, reframes exhausted as a normal
delivery, adopts the upstream label ladder and closing wording, and keeps
the 4/3 confirmation gate separate from the 3/2 delivery floor.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0155nFCgCHtoA7jhSDGmZmMu
Keep the four migrated cases in the queue. Record prior score exposures, distinguish migrated review from a fresh biography audit, and reject exposed or auto-attested cases from blind holdout use.
Co-authored-by: Cursor <cursoragent@cursor.com>
The published six keys stay on the 2026-07-21 freeze. Current-tree sidecar numbers remain diagnostic and unlabeled as release metrics. Remove the unused v2 PILOT_REPORT_PATH.
Co-authored-by: Cursor <cursoragent@cursor.com>
Keep confirmation fail-closed until the public AA set is ready, and rewrite Technique Audit from the attached VedAstro status instead of a hardcoded blocked row.
Co-authored-by: Cursor <cursoragent@cursor.com>
Web answers were thinner than a local Agent calling yinduzhanxing-skill:
theme-subset vargas, no visible audit table, and a prompt that dropped the
invocation contract. Bind the commercial method, compute D1–D60 plus Western
layers, and deliver the same Full-Spectrum checklist.
Co-authored-by: Cursor <cursoragent@cursor.com>
