Once the discriminator training gate is open, only choice cards are asked
and the range card goes out when they are exhausted; targeted lines, their
re-ask and guided windows no longer hold the card or invite more events.
Delivery body says how many choice questions were used instead of the event
fit percent; narration names an excluded cluster instead of "range
unchanged"; a delivered turn no longer carries a collect question.
Offline replay (v4, 3 radii x 2 directions): truth in range 20/20 in every
cell; guided-window injections give the same width in truth and opposite
directions, so red line 1 was revised by product to truth-in-range only.
Skill 10.0.31 -> 10.0.32 (10.0.31 kept as deprecated for pinned cases).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
- Voice: contrast limited to chart structures; drop 「这句我按『……』理解了」; forbid motive/need sentences; yes/no answered first (BUG-1070)
- Follow-up turn (session already holds an answer): ≤200-char direct answer, no skeleton; decided from stored history, no intent regex; tool still runs each turn (BUG-1071)
- Thinking bar first section 「先回答你问的这件事」 (BUG-1072)
- First turn: opener without actions, body headings 盘里支持 / 时间怎么看 / 这周可以做的一件事 only, actions once, ≤900 chars (BUG-1073, D8)
- Checklist is a list to check, not paragraphs to write (T3)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0199rbQDTsUbCVw84wc8BTFe
The marriage evidence card's gender field is the bound subject's own value
(女 / 男, 性别未知 when not filled). The condensed checklist keeps the
gender-unknown rule only when unset; otherwise it names the spouse
significator per strict-workflow-router.md §4 and the repo's gender
interpretation contract (supplement only, core stack gender-neutral). The
methodology cache is keyed by gender so one person's line never serves
another. Tests drive the real getJyotishAgent with a prompt-recording model.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
TASK-consult-evidence-card-v2-20260927 T4 (方案 B). Career, marriage and
wealth hand the answer model a 5-7 line condensed checklist
(frontend/src/lib/consultation-condensed-checklist.ts) in place of the full
strict route plus event-judgment file: must-see items (all on the card),
the three layers (接触 / 结构性机会 / 公开落地; 心动接触 / 关系成对 /
社会法律落地; 挣钱机会 / 真实增长 / 到账变现), the forbidden statements
(5L 大运 ≠ 法律婚, 金星过本命月 ≠ 心动月, Punarphoo 不得写成结婚,
土星回到本命月是观察项), the gender-unknown rule, and the lookups (KP
blocked). The shared baseline still travels; timing / health keep their
full router sections; the router, the event-judgment files, the bound
system method and the skill are unchanged (no skill version change).
Only the consultation tool builds the methodology, so reports and other
surfaces are unaffected (source test).
Single-domain model-visible size, 3 public AA charts x (10 questions +
annual + timing): career 9.8K, marriage 10.5K, wealth 9.9K, annual 11.5K,
timing 11.3K; all <= 12,000.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
TASK-consult-evidence-card-v2-20260927 T3. Card version evidence-card-v2.
- Base section (every card): the engine's D9 summary (D9 lagna, each
planet's D9 sign and dignity, Vargottama, D1/D9 reversals); D9 houses and
aspects stay out.
- Career: + AL (engine pada A1), 10H SAV, SAV of the Jupiter / Saturn
transit signs.
- Marriage: + 5H / 5L, day / night, Punarphoo (observation_only), Double
Transit on 7H / 7L (house-7 run) and DK / UL, Vivah Saham, gender unknown.
- Wealth: 8H / 12H named.
- Annual: the annual Tajika chart verbatim (parameter_sensitive), or
年盘未接入 when the pack is blocked / not attached (never natal data);
houses 1 + running / next AD lords' houses + the year's Jupiter / Saturn
transit houses, with their basis; ingress / station dates only.
- Timing: Rahu / Ketu with ingress dates, Jupiter / Saturn SAV and BAV,
Double Transit conclusions (no degrees); vargas follow the turn's other
domain, D9 when alone.
- Chara Dasha leaves every card; KP is never on a card. The lookup enum adds
karakamsha, dispositor_chains, inter_chart_linkage, argala, moon_transit;
a KP lookup travels with its blocked note.
- System prompt and tool descriptions name the D9 summary, the
parameter_sensitive / observation_only tags, 年盘未接入 and the lookups.
- Telemetry card version / agentVersion bumped to v2 (three-column notes in
the touched tests).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
TASK-consult-evidence-card-v2-20260927 T2. toAgentConsultationContext carries
chart.consultation_native_layers as local_layers.native_layers; toModelOutput
projects d9_summary / punarphoo / vivah_saham / day_night into the natal card
and slow_transits / double_transit / annual_tajika into the timing card, each
through its own nested-key allowlist (as BUG-1054 did for Narayana), so the
existing natal and timing key sets are not widened and the depth / list caps
are unchanged. Pada A1 (the engine's Arudha Lagna) is allowlisted for the
career card's AL.
Golden test: every projected layer equals the engine layer value for value on
3 public AA charts (family, timing and annual routes); the Double Transit
per-house summary line is the only field left out.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
T6 of TASK-rectification-grounding-20260927.
- skill-slice.ts: per-action slice rule as a code constant, keyed on heading
titles (evidence: 「ConversationFocus 与意图承接」「批量证据与日期真实性」, i.e.
§5/§7 of the 10.0.x layout); other actions and any bound Skill without those
headings (9.0.0) get the whole body, so historical Cases still run with their
exact bound Skill (BUG-621).
- The rectification Agent declares providesSkillDiscovery "on-demand"
(rectificationSkillBoundProcessor): no <available_skills> block with a temp
path and no "call the skill tool" system message; getSkill still loads the
bound package.
- Measured on a real Agent + recording model (public AA case, estimate = CJK
chars + other chars / 4): fixed overhead per call 12,002 → 6,471 tokens
(step 0: 8,440 → 2,909). Skill text unchanged; no version bump.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
T5 of TASK-rectification-grounding-20260927.
- rectification-offer-candidates returned the full projection (~100 KB on the
public AA case: *_pre_inference, 31 KB contrast packet, probe lists). It now
returns the compare stripping (offerModelProjection); 100,012 → 28,799 bytes.
- agentVisibleLatestProjection also drops engine_indistinguishable_width_minutes
(audit-only since BUG-593), keeps the verification Markdown once
(skill_verification_report; range_delivery.verification_markdown was a
byte-identical copy) and shows the slim candidate list.
- Receipt fingerprints stay over the full payloads; the case API projection
(latestResultToolProjection) is unchanged.
Contract tests on a real local engine response (public AA chart).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
T2 of TASK-rectification-grounding-20260927 (product decision P1).
The rectification 「重试（重新生成）」 was a blind agent.generate rewrite that
persisted unchecked text (invented range / fit rate reproduced). Removed:
the regenerate route, regenerate-turn.ts, the regeneration agent and its
read-only tool set, the client regenerate action/state and the
regenerating/canRegenerate props. ChatMessageActions renders the regenerate
button only when onRegenerate is passed; ordinary consultation is unchanged.
The DB function regenerate_agentic_rectification_turn is kept (AGENTS §7.6,
retire in a later round). DESIGN.md / VOICE.md updated; source-contract
tests follow with 原值/新值/原因 notes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
T1 of TASK-rectification-grounding-20260927 (recurrence of BUG-588).
- The attempt no longer streams range-changed / rescore-skipped /
compare-failed sentences; the finish whitelists and trims the model body,
then joins the server facts, and emits one final replace equal to the
persisted text.
- P3 whitelist (spoken-grounding.ts): a model sentence with a clock, clock
range or percentage that is not this turn's server fact is dropped whole;
the batch recap stands in when nothing is left.
- record-evidence-batch returns range_after_rescore (post-rescore
credible_range, representative minute, fit percent, delivers_range_this_turn);
the receipt fingerprint stays over the old shape.
- System prompt: range is said by the server; the delivery three sentences
only when the batch says this turn delivers.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
After each natal turn the route logs a separate [agent-observability] event
{runId, agentVersion, evidenceCard: {domains, cardVersion, cardChars,
cardTokenEstimate, citedFieldIds, feedback}}. Cited field ids follow the
research R5 rule (ISO date, degree, planet-in-sign phrase of 8+ chars
appearing verbatim in a finished answer); matched text is discarded. Thumbs
are client state only today, so feedback is "none"; storing thumbs per turn
needs a table and is left to a follow-up. The strict schema has no user,
session, question or answer field.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
read-consultation-evidence returns one closed-enum section (a formal varga,
research/extended vargas, a Western layer, yogas, Ashtakavarga, Shadbala,
transits, Chara Dasha, arudha, karakas, KP, gulika, kakshya, mahadashas,
thematic evidence) from this request's finished calculation, never
recalculates, answers unavailable on a cache miss and refuses a second call.
The receipt records the step and the write row shows 「正在多看一眼：…」.
A lookup after answer text went out keeps the released text whole: a verbatim
restart is dropped as it arrives (40-char confirmation), a continuation is
kept, and settlement still reads the step that wrote the answer; the lookup
runs on the answer clock without resetting it. A length continuation carries
the lookup result with the card. BUG-1059: the visible-text transformer's open
clause is settled at each tool call, so unpunctuated narration no longer
leaks into the answer.
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
New frontend/src/lib/consultation-evidence-card.ts ports the research
CARD_SPECS: base section (ascendant, house signs, placements with degrees,
functional benefics/malefics with lordship, Vimshottari MD/AD/PD with dates,
Narayana md/ad/pd) plus a section per domain; values copied verbatim from
the projection or engine context, gaps listed, never filled. The tool now
returns toModelEvidenceView: status, evidence_contract (policy, blockers,
layers, limitation), claim_cards = the card, evidence_card meta (D4 line,
supplementable sections), rectification, methodology, domains; the audit
table, spectra, must_use_layers and presentation stay server-side and the
single-domain consultations copy is gone. 「本轮技法」 rows are unchanged.
Golden tests over three public AA engine captures.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
parents (D12, 4/9 houses, Sun/Moon) and children (D7, 5th house, Jupiter,
PK) are plan/card domains with Chinese labels and the brief's aliases. They
send the family route contract to the engine and are rejected as a stored
session theme, so neither Python nor the chat_sessions.theme CHECK changes.
Methodology reports no strict checklist for them; must-use layers follow the
plan domain. Tests with 原值/新值/原因 where assertions changed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
timingKeys now admits current_dasha.md/ad/pd (sign, lord, years,
start_age, end_age), remaining_years and pratyantar_dasha_timeline. Depth
and item caps unchanged. Golden regression over three public AA engine
captures asserts values, not key presence.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
The natal loop's step after run-jyotish-consultation saw the evidence, but
its text was drained and a second, blind compose stream (history + question
only, empty findings) wrote the user-visible answer. Remove compose,
interpret and the drain; keep the loop's own final-step text.
- stepScopedAnswer: per-step holding; text of a step that calls a tool is
dropped, so narration around tool calls never reaches the answer
- writing shape (opener + four headings) moves into the user turn
- length continuation receives the calculation result; Pass 4 whole-answer
reject retries through retryForAnswer with the rewrite hint
- createConsultationRunClock: tools keep the 110s tool phase; the loop is
handed to the 70s answer clock when the calculation result arrives
- settlement judges the step that wrote the answer (BUG-1051 kept)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
The tool loop and the answer-writing stream shared one 110s AbortSignal.
Mastra 1.50 does not throw on abort: it emits an abort chunk and
finish(tripwire) and closes normally, so a half-written answer reached
onComplete, was charged and persisted as completed.
- Compose, length continuation and answer retry run on a 70s answer clock
started on first use (worst case 110s + 70s = 180s; maxDuration 240).
- Settlement requires finish=stop from the stream that wrote the answer;
abort/tripwire, content-filter, tool-calls, other/unknown/error or a
missing finish with visible text ends as answer_truncated (cancel, no
charge). The abort chunk records an abort runtime step; a cut stream no
longer flushes its dangling Pass 4 sentence. length still continues.
- [agent-observability] gains composeFinishReason, composeAborted and
answerVisibleChars (enum/boolean/count only).
- Regression tests use a real Mastra Agent over a fake model; the BUG-305
hand-thrown DOMException fixture is kept with a three-column note, and
eleven fixtures gain the finish(stop) chunk real streams always carry.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
- BUG-1049 (recurrence of BUG-504): the opening stem carries six examples and
an example answer again and is server-owned on the zero-evidence opening;
the body is two plain sentences (no 大运/盘面/代表分钟/精确到秒, no year, no
question). Stem de-dup compares whole sentences / near-equality instead of a
12-char prefix, which had deleted the body's examples sentence since
aa7ccb30 (BUG-604) + dd8f35f7 (BUG-648).
- BUG-1050: plain step labels; a finished step label shows once and
「已完成 N 步」counts shown rows; failed rows read 「…未完成」 from the
in-progress wording.
- Skill 10.0.30 -> 10.0.31 (OpeningPolicy); 10.0.30 kept as deprecated.
- VOICE / DESIGN / CHANGELOG / BUG_HISTORY / PROGRESS / real-device checklist
and screenshots.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017eEAG8HD3mm8gsKXgk8uU8
Unify reader cleanup rules, lock writer table guards, and register the exact fictional timestamp collision. Preserve existing ordinary-report safety contracts and source-data gaps.
Validation: report Node 165/165, final safety 29/29, Python 101/101, Chrome 28/28; both PDFs retain all 130 rows. Full Node 3704 tests with the same 91 baseline failures. Privacy test: 62 passed, 1 failed due to 17 protected-file READ_ERRORs; not a green gate. Build, DB, manual checklist and controlled-login gaps remain documented. User explicitly authorized staging push with these gaps disclosed.
Co-Authored-By: Claude Code <noreply@anthropic.com>
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.
Unify minute and block cache identity, keep unverifiable historical results read-only across server tools and write entrypoints, and aggregate completed receipt sources chronologically through a compatible function migration.
Co-Authored-By: Claude Code <noreply@anthropic.com>
Implement BUG-984 F2 option A and reproduce mixed completed identities through real tools. Stop at the required SQL authorization boundary; F1/F3/F4 remain pending.
Co-Authored-By: Claude Code <noreply@anthropic.com>
Keep full consultation tool contracts unchanged. Persist short replies and refund the original reservation atomically while recording actual model usage. Verify Linux frontend 3566/3566, database 40/40, Static home and gzip +0.0493%.
Co-Authored-By: Claude Code <noreply@anthropic.com>
Homepage no longer waits on /api/onboarding for six starter questions.
Natal answers show the spoken layer first; the Level 2 skeleton sits in a
collapsed 完整分析 block. Wide tables scroll sideways on a phone. Form
inputs are 16px so iOS does not zoom on focus.
Keep targeted existence questions as A-D cards. Recover collect-schema
stock by question-id prefix, surface the stem when persist fails, and
send spoken targeted existence to the repair exit instead of a naked prompt.
Hide the button once both personality cards are answered, persist the second card in the same opt-in round, and freeze Skill 10.0.26.
Co-authored-by: Cursor <cursoragent@cursor.com>
Same-id focus returns instead of making the model retry; kind_hint targetKind is ignored. Failed receipts keep a redacted original error and known codes. Host fallback recap no longer duplicates dates.
Co-authored-by: Cursor <cursoragent@cursor.com>
Occupation collect was wiping year-month into an unscored note, and idle gap copy never joined the evidence turn.
Co-authored-by: Cursor <cursoragent@cursor.com>
Stop domain-wheel collecting and age-band years in prompts. Ask until the training gate, then discriminate until convergence, then deliver a range plus a concrete follow-up. Reserve holdout only with four dated events.
Co-authored-by: Cursor <cursoragent@cursor.com>
Ask finance and health like other domains, put year-cued collect questions first, and let the classifier distinguish decline vs skip. Skill 10.0.22.
Co-authored-by: Cursor <cursoragent@cursor.com>
Host only says 记下了 after a real write; Mastra schema rejections fail closed. Ledger year keys no longer drop quality probes, dual-dasha agreement is per cluster, width uses cluster span, and public house tables follow the inference minute.
Co-authored-by: Cursor <cursoragent@cursor.com>
Evidence turns that already recorded a batch no longer fail the whole run when the model emits no text; unchanged ranges now name which clock spans lead or lag, and the timeline no longer says 收窄.
Co-authored-by: Cursor <cursoragent@cursor.com>
Thinking models were spending the first natal step on skill_read or a Level 2 draft, so home chips such as 家庭 never satisfied the calculation contract.
Co-authored-by: Cursor <cursoragent@cursor.com>
Homepage「深入看今日」was rewritten to natal 综合, then each compose slice burned its only step on a tool call, so the answer stayed empty. Pin the route theme, write the three daily sections with tools disabled, and filter thinking by sentence so English word-salad cannot leak.
Co-authored-by: Cursor <cursoragent@cursor.com>
