- BUG-1729 (P0): the execution receipt capped methodologySections at 12 while
the loader allows 32; a wealth turn delivers 14 and every receipt failed.
The cap now lives in consultation-methodology-limits.ts and both read it;
a sweep of all 2509 domain combinations x 3 questions passes the receipt.
- BUG-1730 (tooling): the offline model-input manifest moves to
tests/helpers (scripts must not import tests, BUG-1140); the fact-checker
reads the timing route's nested slow_transits.
- BUG-1731: the card carries the engine's calendar_year (both Tajika years
and the splitting return); runtime contract rule A-08 (her Q2).
- BUG-1732: narayana_method_status names the undetermined / disputed step
(seed, years, dual_lord, ad_direction) from the engine, "unknown" when it
cannot; B3/B4 recorded as open; chat_use single_system_candidate /
no_judgment; her Q1 sentence in the contract; B-02 rewritten.
- BUG-1733: the conclusion's first sentence answers the question; the
frozen time range follows it (her Q5).
- Golden recaptured with TTL=0 (method_status; VedAstro offline reason is
environment-dependent).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
- BUG-1725: evidence card Arudha rows carry pada_house_from_lagna and
lord_natal_house plus arudha_columns; her phrasing rule
「A2 位于水瓶；其主星土星位于本命第 2 宫」 is in the runtime contract.
- BUG-1726: methodology delivers her files in her three layers (ruling
2026-11-07 Q3): router, status contract, output contract (skeleton
§5/§6) and §0.4 every turn; §2.3, the matching event-judgment branch,
the transit guide and route-needed sub-skills per question; maintenance
files never. methodology.delivery records layer and reason. Removes
the dead sharedConsultationMethodMarkdown.
- BUG-1727: sub-skills follow route techniques through her
skill_family_router.json triggers (ruling Q4); question words are
auxiliary; product domain table kept as labelled floor;
specialized-jyotish reachable as technique paragraphs plus its
boundaries; skill_read reads allowlisted skills/<family>/ files.
- BUG-1728: CHAT_METHOD_BOUNDARIES, natal instructions, MEVG policy and
the shared reading move to references/product_runtime_contract.md
(loaded by block name, each rule annotated as a product rule); her Q5
rulings added; current order and a list of rules that may cross her
writing floor recorded for her.
- Offline model-input manifest mode in consult-model-input-manifest.mts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
- Backtest harness writes, per run, the exact model request bodies (no
headers or key), the computed card and a manifest: system prompt
sections with char counts, tool calls and results, and a truncation
check of the transmitted card (parses in full, equals the computed card,
strings at the 800 cap, arrays at projection caps or shorter than the
engine output, card gaps). --no-manifest turns it off.
- consult-answer-factcheck.py extracts checkable claims (regex, plus
optional model extraction into JSON only) and verifies every claim
deterministically against the card: matches / contradicts / not_on_card
/ status_mislabeled; per-layer transmitted / named / cited-correctly /
audit status with "Used but never cited" flags; date precision,
conclusion position and grade-rule checks; --report confirms answers
written from a skill report.
- Professional voice (structure only): a short 结论 block first; point and
lord facts only from the card field, one value per point; status words
copied (the Yoga line no longer says 「只写 yogas.hits 里成立的」); a dasha
level the card does not give is 未交付; dates are 名义计算边界 unless the
birth time is rectified; each grade cites its rule and missing items.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyn7Hj3LCW3BUs5bZ8
Home chat surface, onboarding assessment, rectification cases API, Mastra
rectification tools and admin telemetry are gone; 生时纠正 is the standalone
/rectification page. Old birth_time_rectification sessions are hidden from
lists. Consult offer card links to the page. Copy renamed to 生时纠正.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXozbkyFNqbu8pVEtZ7W
Two earlier attempts asked the model to reason from the age (a voice clause
with no birth date; an age_years field it ignored). The server now decides:
when the native is ELDER_PARENTS_AGE_THRESHOLD (65) or older at the request's
UTC+8 calendar date, the parents, family and general sections carry one plain
sentence right under the label, parents_life_stage, with the age in it. Below
65, or with no birth date, nothing is added. The birth date, time and place
still never reach the model; no prompt rule changes.
The biography backtest reads each figure's real birth date from the golden
instead of 1950-01-01 and passes its request time.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyeZrYn6ryNcSe5Y2Y
The parents answer read as house-by-house fact chains with a gloss on every
house, opened with "both lines show pressure" instead of the pattern asked
for, and copied the example's 「实际是哪一种，告诉我一句」. The model was
satisfying ~13.9k characters of writing rules stated three times (voice,
English contract, user turn) plus copyable Good/Bad examples.
- VISIBLE VOICE is now the single statement: follow the user's own
structure, open with the overall picture, explain why each placement
means what it means, gloss terms briefly; then the red lines only.
- Contract and first/follow-up user turns point at it instead of
restating it; all example answers removed.
- AFFLICTION_RANGE_RULE keeps the red line (no biographical verdicts, no
reverse reassurance) but drops the forced range and closing ask; the
question's own premise counts.
- Contract tests cap the voice at 2600 characters and lock every red line
the old tests guarded; backtest script gains --question.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXoyeZrYn6ryNcSe5Y2Y
Consult, rectification turns, and report generation stop before a credit reservation when the product caps are hit. A full heavy-compute gate waits up to 8 seconds, and one consultation retries that 429 inside the same request.
Co-Authored-By: Grok (agent loop)
The dry-run script lived under frontend/scripts and read a test fixture, so the web image build would type-check a file the image does not copy. Move it to frontend/tests with a non-test filename. The switch default is unchanged.
Checked: guard plus answer-effort tests 5 pass, 0 fail, 0 cancelled; probe --dry-run exit 0; tsc exit 0; lint 0 errors and 123 warnings.
Co-Authored-By: Grok (agent loop)
The switch is off unless CONSULT_ANSWER_REASONING_EFFORT is set, so today's request body stays the same. A set value is sent as top-level reasoning_effort only on the answer step. The real A/B did not run: there is no model key.
Request-body probe, the named frontend tests (114 pass, 0 fail), tsc, and lint (0 errors) were run.
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
The cache key now includes the request date, and files older than two days are removed. The model card keeps values only. Parashari aspects are copied from the engine. Saving a chart warms the packet without blocking the response.
Biography severe conflicts rose from 1 to 3, so this is not accepted for merge. Not pushed.
TASK-rectification-in-chat-step1-20261006, direct execution.
- The homepage 生时校正 pill and the text trigger (isRectificationHandoffQuestion)
are gone. The consult agent decides (RECTIFICATION_OFFER_RULE, one
definition) and calls offer-birth-time-rectification; the server decides
whether a card may show (self, product on, no case in progress,
feature_pricing price). The offer is stored with the answer and travels on
run.completed; the card hangs under that answer.
- Opening from the card binds the offer to the case in the consultation's own
messages (no table change); GET/POST /api/rectification/cases/[caseId]/source
find the source and go back: one divider (time changed or not), earlier
answers captioned 按校正前的时间, and a 按新时间重新看 button that sends only
when clicked. Auto-return after the post-adopt check or a terminal case, only
for work done in this view.
- Model history skips the divider and notes answers written with the old time.
Full suite 4987 / fail 24, identical to 6ec8efd5 (4973 / 24), no test name
lost; tsc 0; lint 0 errors; build keeps / Static, first-load gzip +0.20%.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
- Evidence-card and biography-backtest goldens regenerated with the capture
scripts (PYTHONHASHSEED=0, cache TTL 0; byte-identical twice); only the
functional-role fields change.
- CAREER_FIELD_ASK_RULE: A10 / AL / Sun / 10th lord in 4/6/8/12 say
resistance or cost only, never "not on stage / go deep", and houses are not
translated into fields.
- AFFLICTION_RANGE_RULE: no portrait of an afflicted person from the lord's
house or a benefic occupant ("managing / arranging / cares about your
future"), actions do not presuppose reaching them, one ask per answer.
- Examples that paired "4th lord in 10th" with "her attention is on your
future / close" now state the line is unafflicted (Monroe's chart had that
exact placement and copied it). Retrograde line also bans "drags on".
- Backtest script: several follow-up domains in one run.
Changed assertions carry 原值 / 新值 / 原因 in the test comments.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
- natalAnswerShapeBody: NATAL_SECTION_RULE replaces 「这个人是什么样、你们怎么相处、
事情会怎么走」; the shared voice quotes the same constant (BUG-1164)
- product-voice / VOICE.md: no 逆行 in examples, parents example leads with the
affliction and asks to confirm, married-presupposing Good example removed (BUG-1165)
- follow-up shape: a user correction is acknowledged, never defended (BUG-1166)
- career examples carry no job shape (升职 / 专业能力 / 做深现有工作) (BUG-1168)
- biography backtest golden regenerated with the real engine so the card carries
brief 1's aspects, combustion, yoga domains and saturn_from_moon_house;
backtest script gains --correction follow-up turns
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
TASK-consult-no-presupposition-and-backtest-20261001 T4 infrastructure only
(T1-T3 untouched). New files only, so it merges cleanly with the parallel
evidence-card brief.
- capture_consult_biography_backtest_golden.py: same handler/body/trim as the
evidence-card golden; nine public AA charts x parents/marriage/health/career.
- consult-biography-backtest-golden.json: real engine output, byte-reproducible.
- consult_biography_backtest_rubric.json: facts with sources, must_not,
expected_signals per figure x domain.
- consult-biography-backtest.mts: runs the product's real agent, tools, card,
methodology, user-turn shape and streamAgentResponse; deterministic checks only.
- Baseline on 9b937c4a: 39/72 severe biography conflicts; all five audit cases
reproduce in both runs.
BUG-1170 is registered when brief 3 lands.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
The staging publish job built the web image without frontend/tests, and
next build type-checks every .ts under frontend/, so the harness importing
../tests/helpers/postgres-fixture.ts failed with TS2307. The harness depends
on the test fixture, so it now lives in frontend/tests/research; the Python
driver points there. A new test fails if src/ or scripts/ import tests/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
- open_agentic_rectification_case_v2 (12 args) becomes SECURITY INVOKER: the
immutable-skill ACL reconciliation leaves EXECUTE on the 11-arg open only to
service_role, so the definer wrapper hit 42501 on every homepage/new open.
- The pending-opening read moves to owner function
agentic_rectification_opening_pending_v1, granted to service_role only.
- History test and persisted replay harness append turns through the V10
request-idempotent overload; the V9 overload is revoked from service_role.
- Opening test fixture adds the required birth_time_source.
- Segment migrations renamed to 20261001* so they sort after staging's
20260930* migrations on both fresh and existing databases.
- Stale Windows replay replaced with the production-path replay (M1/M2 equal
to accepted research, implementation_identity included).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
Review-only snapshot for BUG-1115 through BUG-1117; not merge-ready. New opening and append-turn PostgreSQL permission failures remain blocked. Persisted joint replay has zero completed questions; segment ordering remains off by default. The existing offline replay JSON is retained stale and unchanged after a denied overwrite, including its CRLF line endings. Browser/provider validation and final serial gates remain pending. No deployment, role permission changes, or staging/main push.
Co-Authored-By: Claude Code <noreply@anthropic.com>
An inline classic script after the stylesheets waited for them to download;
on staging the reads left at 2.4-3.1 s. Measurement script gains
--css-delay and --ua iphone.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Progress with before/after, T4 evidence (supportsImmutableAssets has no
effect under self-hosted standalone) in BLOCKED.md, device checklist,
changelog, board row to 待验收.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE
Rectification answers now advance chat_sessions.updated_at (BUG-704).
A ?c= id missing from the loaded page is fetched before anyone may call
it deleted (BUG-705). The range card no longer has a post-card tie-break
button; live questions leave the card visible with adopt locked
(BUG-706/708). Spoken copy bans 相对支持度 (BUG-709).
Task docs assigned 700-704; qizheng already took 700-703.
The streaming and settled versions of the trailing assistant reply were
two components, so settling unmounted one and mounted the other and the
entrance tween replayed over text the reader was already on. One
LatestAssistantEntry now owns that row under a single key, and the
history list excludes it. The CSS entrance keyframe that doubled the
GSAP tween is gone and the tween matches the documented 160ms. The step
timeline no longer remounts on settle: it is a button-controlled
disclosure with a 180ms grid-rows transition, the reader's own toggle
wins over the live default, and an in-flight request with no events yet
shows a queued row instead of an empty shell.
BUG-474 BUG-475
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JUei7K13cYxLHE3Axe4A45
Settled messages no longer rebuild on every token, Inter is actually
requested, and admin routes drop the 33 KB chat stylesheet. Root
force-dynamic is gone so public shells can prerender without changing
the no-store Cache-Control contract.
Co-authored-by: Cursor <cursoragent@cursor.com>
ClaudeandClaude Opus 5.5