The chair (Dafna/chaim) reviewed and approved the full corpus-refresh proposal
generated by scripts/regenerate_voice_fingerprint.py and committed it to the live
voice-fingerprint — satisfying the INV-LRN1 hard chair-gate (a human approved;
the script only ever proposed). +81/-15 lines: all the deletions are reword-and-
expand of existing lines (no curated insight lost), the additions are style/method
refinements grounded in the 56-decision corpus (INV-LRN5 — no case substance):
- §1.2 conditional-acceptance as an "anחנו" verb (מקובל עלינו vs נציין)
- §1.3 ruling variation "ראשית, צודק [צד]... אלא ש"
- §1.5 emphasis always annotated "(הדגשת הח\"מ)"
- §1.8 bounded appraiser-restraint (professional plane only; legal plane = no deference)
- §1.9 inverse twin "ניתן היה לסיים... אלא ש"
- §1.11 quantitative anchoring as a reframing device (cross-type)
- §1.12 measured block ratios (background 17 / claims 27 / discussion 42 / summary 14)
- §1.13 faithful rendering of a party's internal reasoning
- §2.2 two-sided pendulum for the restraint paragraph + expanded party-verb catalog
- §4.8 new template "ניתן היה לסיים — אלא ש"
- §5.x agent instructions updated to reference the new patterns
Invariants: INV-LRN1 (chair approved before commit — the gate held), INV-LRN5
(style/method only, no substance).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The voice-fingerprint (docs/daphna-voice-fingerprint.md) is the PRIMARY abstract
style channel the writer consumes (07-learning §0.2 channel A), but it was
hand-authored and never regenerated from the growing corpus — the "voice" the
writer adapts stayed frozen while finals accumulated. PR #345 unfroze the
structural half (corpus-measured ratios → writer); this is the prose half.
scripts/regenerate_voice_fingerprint.py reads the corpus (style_corpus stats +
a bounded block-level style_exemplars sample + measured section ratios) and asks
Opus (local) to PROPOSE refinements/additions to the fingerprint — STYLE / METHOD
/ VOICE / LEXICON only, grounded in corpus evidence, never case substance
(INV-LRN5), preserving curated insights rather than replacing them.
CHAIR-GATED, HARD (INV-LRN1): the prose profile is voice-knowledge, not a
low-risk style rule, so it does NOT auto-flow. The script writes a PROPOSAL to
data/curator-proposals/ (the existing curator-proposal surface) and NEVER
overwrites the live fingerprint; the chair reviews and hand-commits accepted parts
— the same manual gate the fingerprint already uses. Runs on the host (claude CLI).
Reuses claude_session + measure_corpus_ratios + the curator-proposals dir (G2 —
no parallel synthesis or gate). SCRIPTS.md + spec §0.2 updated.
Invariants: INV-LRN1 (hard chair gate for voice-knowledge prose), INV-LRN5
(style-only synthesis), G2 (reuses existing synthesis inputs + proposal surface).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The voice-learning panel produced 81 vetted style lessons — all stuck at
review_status='proposed' with 0 ever approved — so the writer (which reads only
'approved') received NONE of them. The system captured learning but never let it
flow. Chair decision (2026-06-28): a GRADUATED gate by content risk.
- STYLE lessons (categories style/structure/lexicon/tabular) the 2/2 panel kept →
created as review_status='approved' → flow to the writer immediately, reversibly
(chair vetoes in /training). The deepseek+gemini panel only emits style_method
and only on 2/2 consensus, so this is exactly the gate's criterion; substance is
already filtered out and skipped, and routes through the strict halacha gate.
- SUBSTANCE (halacha/precedent/fact) stays a HARD chair gate — unchanged.
style_lesson_panel.py: _review_status_for(category) sets the gate explicitly
(approved for _STYLE_CATEGORIES, else proposed); the apply loop passes it to
db.add_decision_lesson (which already accepts review_status — no DB change).
Spec: INV-LRN1 rewritten as the graduated gate (hard for substance; reversible
auto-flow for consensus style — still "under user control" via veto, per
NCSC/CEPEJ); §0.6 updated to match.
A one-time backfill of the 81 existing panel style lessons to 'approved' runs
separately (DB, post-merge).
Invariants: INV-LRN1 (amended — graduated), INV-LRN5 (style-only, substance never
auto-flows), G10 (human control preserved as reversible veto).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
תיקוני-מבנה ללולאת-הלמידה (TaskMaster #157):
1. כרטיס ה-curator (/training טאב "אוצֵר") ספר decision_lessons WHERE source='curator'
— ערך שאף מסלול-קוד לא כתב → תמיד 0. עוצב-מחדש (דרך שער-העיצוב Claude Design, אושר)
להציג ביושר את שלושת ערוצי-ההזנה לכותב: דיסטילציה→appeal_type_rules (180, זורם),
פאנל→decision_lessons (81, ממתין), אוצֵר→source='curator'. get_curator_stats שוכתב.
2. drift ספ↔סוכן: 07-learning.md §1.1 + INV-LRN3 קבעו שהאוצֵר רושם ממצאים כ-decision_lesson
source='curator', אך הסוכן כתב comments בלבד — הממצאים האיכותיים אבדו. נוסף כלי-MCP
record_curator_findings + §A.5b ב-hermes-curator.md (read-only נשמר; הצעה מגודרת-שער).
3. get_recent_decision_lessons(limit=15) חתך בשקט — נוסף WARN על מה שנחתך (חוקה §6);
הפתרון האמיתי = סינתזת-לקחים (TaskMaster #158).
Invariants: מקיים INV-LRN1/G10 (שער-יו"ר), INV-LRN3 (לכידה מובנית), INV-IA2/IA5 (מקור-אמת
יחיד), G12 (leak-guard עובר), G2 (אין מסלול מקביל — אותם מאגרים). פער-מימוש פתוח מתועד:
§A לא רץ אוטומטית על mark-final (pipeline-wake exits) → נדחה לתכנון נפרד.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Documents the prospective held-out methodology shipped in PR #337: why a clean
retrospective held-out is impossible (lessons stored universal/untagged → no
leave-one-out), and how the system instead captures a clean generalization
datapoint at final-upload — before this case's lessons are folded — via a
style_distance snapshot + lesson-pool size into style_distance_history,
surfaced by GET /api/learning/style-distance-history.
Adds §0.7, wires step [7] MEASUREMENT and INV-LRN4 to it. Notes the
change_percent style/content confound and the one-time clean window.
Invariants documented: INV-LRN4 (this is its trend surface), G2 (reuses
style_distance + appeal_type_rules — no parallel metric path).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Inline # comments in gitignore are not supported — they were silently
breaking three patterns (data/checkpoints/, data/adapter-migration-state.json,
.claude/agents/.generated/). Moved comments to their own lines and added
missing entries for runtime dirs (data/audit/, data/logs/, etc.) and
temp files (.interaction_tmp.json, .design-build/, .taskmaster bak files).
Also tracks previously untracked legitimate files: scripts, tests, docs,
skills references, .env.example, taskmaster templates.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
chaim directive 2026-06-20: no halacha queue ever; trusted = only what a chair
actually cited, in context; corpus grows by citations; few-verified-now is fine.
nli-audit: nli_unsupported flag is 97% false-positive (29/30) — the 'noise' was
imaginary. Two layers: raw background (auto, no gate) vs verified (auto-canon from
citations). +06 indexed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
25 primary sources, 21/25 claims verified (3-vote adversarial). Verdict: keep
all principles, rank by precedent-level citation centrality at retrieval time
(not a destructive holding-level cull — that's where automation provably errs),
gate human review near-zero via conformal/selective-prediction.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ועדת-ערר decisions (source_kind=internal_committee / source_type=
appeals_committee) are persuasive authority — they do not bind another
committee, nor the committee itself. The stored is_binding column wrongly
defaulted to True across the FastAPI form + service/db layers, so 46 of 92
committee rows were marked binding and got the BINDING halacha-extraction
prompt. Authority is structural for this source (INV-DM7) — normalize at the
source (G1), not trust the input.
Changes:
- db.create_internal_committee_decision: coerce is_binding=False (structural)
- db.create_external_case_law: coerce False when source_type=appeals_committee
- db.update_case_law: coerce False when a patch relabels source_type=
appeals_committee (Gemini reclassification path)
- internal_decisions.migrate_from_external_corpus: set is_binding=FALSE on
the external→internal reclassification UPDATE
- service + FastAPI-form defaults: True → False for the internal path
- SCHEMA_V42: backfill legacy committee rows (is_binding True→False) +
CHECK constraint case_law_committee_not_binding_check so a binding
committee row can never be written again ("so it doesn't recur")
- spec: X8-field-provenance aligned to INV-DM7
Verified against live DB in a rollback transaction: 46 rows flip to 0;
court_ruling (239) + cited_only (31) unaffected; binding-committee INSERT
rejected by the constraint.
Invariants: INV-DM7 (authority ⊥ rule-type) · G1 (normalize at source) ·
G2 (single source of truth, no parallel path).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Phase C — scripts/cull_principles.py: re-adjudicates every existing 'original'
principle with the SAME panel regime (panel_keep_score → classify → apply_cap),
reversible (CSV backup + rejected canonical recoverable), usage-throttled.
panel_extraction.panel_keep_score + apply_cap (shared, G2). Dry-run on 3
decisions: 37→15 survive.
Phase D — services/principles.py: source-derived label הלכה (binding court) /
כלל פרשני (committee) / עיקרון (persuasive); umbrella עקרונות משפטיים. Wired into
canonical_halacha_get/list (principle_class+principle_label). UI string changes
deferred to the Claude Design gate. spec INV-LRN7; SCRIPTS.md; 7 new tests; 428 green.
Phase E needs no new code — synthesis already targets pending_synthesis, which the
cull leaves only on survivors (rejected canonicals → 'rejected').
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Grounded (INV-AH) multi-instance synthesis with drift guard + chair gate
(pending_review, G10). Single path used by backfill, MCP tool, nightly drain.
HELD from production run pending the principles-redesign (rename+cull, #152).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document the Paperclip event-shape quirk that silently disabled user-comment→CEO
routing: `issue.comment.created` carries the issue id in `event.entityId`, not
`payload.issueId` (payload has commentId/bodySnippet/reopened). Includes the
empirical log from case 8124-09-24, the issue_assignee_changed cancellation
chain, the fix (plugin PR #2), and verification.
Invariants: upholds CLAUDE.md "ניתוב comments דרך CEO"; G12/X15 (fix lives in the
platform-port shell, not the decision skills). Docs-only.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
נוהל-יו"ר (2026-06-11): מבנה מספר-תיק = <סידורי>-<חודש>-<שנה>, ואורך הסידורי
מקודד את סוג-ההליך — 4 ספרות = ערר, 5 ספרות = בל"מ. הספרה הראשונה ממשיכה
לקבוע תחום בשני האורכים (1→רישוי, 8→היטל, 9→פיצויים). הכלל חד-כיווני:
5-ספרתי הוא תמיד בל"מ; 4-ספרתי אינו מחייב ערר (בל"מ-מורשת מזוהה מהנושא).
הבאג שדיווח עליו היו"ר: חיפוש פסיקה-חסרה לפי מספר-תיק החזיר 404 על כל ערך
שאינו תיק קיים — שבר את הטבלה תוך כדי הקלדה ועל מספרי 5-ספרות.
תיקונים:
- web/app.py: GET /api/missing-precedents — מסנן case_number שלא תאם תיק מחזיר
רשימה ריקה (200), לא 404. סמנטיקה תקינה ל-collection-filter.
- missing-precedents/page.tsx: debounce (350ms) על שדות-הסינון — קוורי אחד
אחרי שמפסיקים להקליד, לא אחד לכל הקשה.
- practice_area.py: regex סידורי \d{4}→\d{4,5}; case_serial_digits() +
is_blam_by_number() (5⇒בל"מ); derive_subtype_with_blam ו-derive_proceeding_type
מזהים בל"מ גם מ-5-ספרות (בנוסף לנושא). callers: cases.py, internal_decisions.py.
- proofreader.py: דפוסי חילוץ-שם-קובץ \d{3,4}→\d{3,5}.
- web-ui: practice-area.ts (מראָה ל-backend), schemas/case.ts (regex
serial-month-year, 4-or-5 ספרות, superRefine 5⇒בל"מ), placeholder בוויזרד.
- תיעוד: docs/spec/X1-identifiers.md §1א + legal-ai/CLAUDE.md.
Invariants: מקיים G1 (נרמול-במקור — ספרה ראשונה כמקור-אמת יחיד לתחום),
G2 (מסלול-סיווג יחיד, אין כפילות), INV-DM/X1 (מפתח קנוני + proceeding_type).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
מחלץ את דיסציפלינת מניעת-ההזיות לבלוק קנוני אחד (docs/anti-hallucination-gate.md)
ומחיל אותו אחיד על כל הסוכנים — במקום שכל סוכן ימציא אותה מחדש ad-hoc (G2: בלי מסלולים מקבילים).
5 טכניקות, כל אחת מעוגנת במקור מקצועי:
- AH-1 עיגון-מקור (אפס ציטוט מהזיכרון) — Stanford RegLab/Magesh JELS 2025 (כלי-RAG משפטיים הוזים 17-33%)
- AH-2 quote-or-retract + AH-3 abstention — Anthropic Reduce-hallucinations
- AH-4 תיוג-ודאות — NIST AI RMF GenAI Profile + RAGAS
- AH-5 Chain-of-Verification — Dhuliawala et al. arXiv:2309.11495
הפצה DRY: הפניה ב-HEARTBEAT.md (נקרא ע"י כל סוכני Paperclip) + שורה אחידה בבלוק
'קרא לפני פעולה' של כל 8 הסוכנים, עם הערת-יישום לכל תפקיד (writer=read-only, qa=אוכף,
proofreader=אל תתקן לכיוון מונח משפטי, exporter=אפס מהות חדשה).
בנוסף: legal-ceo.md מקבל ידע על 'שטן מליץ (Gemini)' עם מדיניות on-demand טהורה —
לא בפייפליין, מופעל רק לבקשת חיים/דפנה, הפלט=לידים ליו"ר (לא לכותב, human-in-the-loop).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
סוגר את הפער שלולאת-צמיחת-הקורפוס (07-learning §1.3) הוגדרה אך לא חווטה: מסלול
/final/upload הכניס רק לקורפוס-הסגנון, וההכנסה ל-case_law הייתה best-effort
שקטה שנכשלה כש-chair_name ריק.
web/app.py — /api/cases/{case}/final/upload עכשיו, סינכרונית:
- קובע chair_name דטרמיניסטית (תיק → ברירת-מחדל-ועדה לפי prefix; לעולם לא ריק →
אילוץ case_law_internal_chair_check תמיד מסופק). לא נשען על חילוץ-LLM —
להחלטות שלנו היו"ר ידוע.
- מכניס את ההחלטה ל-case_law כ-internal_committee (תמיד, לא best-effort) →
ברת-ציטוט בהחלטות עתידיות. מטה-דאטה נוסף מועשר אסינכרונית (Gemini).
- מחלץ את הציטוטים שההחלטה מצטטת (extract_internal_citations), ו**מסמן
אוטומטית** כל ציטוט שאינו בספרייה כ-missing_precedent (open) — dedup מול קיימים.
- התוצאה מוחזרת ב-response (enrolled/linked/missing_flagged) — לא נבלעת בשקט.
הציטוטים-המקושרים מזינים את לולאת-ה-corroboration (X11) — תוקן הניתוק שבו
החלטות שלנו לא היו ב-case_law ולכן לא חיזקו הלכות.
web-ui — toast מציג "נוספה לספרייה · N ציטוטים · M חסרים סומנו".
ספ: 07-learning §0.6 עודכן. אומת ידנית על בל"מ 8126-03-25 (15 קושרו / 6 סומנו).
Invariants: INV-LRN4, X11; G2 (יכולת חסרה, לא מקבילה); feedback_silent_swallow
(כשל-הכנסה צף, לא נבלע); DM7 (סמכות נגזרת).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
מוסיף מסלול ייעודי לקליטת ההחלטה החתומה של היו"ר, ומפעיל אותו דרך שני
שלבים אוטומטיים מדורגים עם פאנלי-סוכנים (אוטו-אישור + אסקלציה ליו"ר).
Backend (web/):
- POST /api/cases/{case}/final/upload — קליטת final חיצוני: שמירה קנונית
(סופי-{case}.docx + עותק קורפוס-סגנון תחת case_number מלא כדי שבל"מ לא
יתנגש עם ערר באותו מספר), פתיחת draft_final_pairs (final_received). לא נוגע
ב-active_draft ולא מריץ retrofit (נבדל מ-exports/upload ו-mark-final → לא G2).
- POST .../final/run-learning + .../final/run-halacha — שלבים מדורגים שמעירים
worker מקומי (claude/DeepSeek/Gemini מקומיים בלבד) דרך הרחבת
wake_curator_for_final עם param task=learning|halacha.
פאנל-סגנון חדש (scripts/style_lesson_panel.py): שני שופטים (DeepSeek+Gemini)
על-גבי דיסטילציית-ה-Opus; הסכמה 2/2-keep → decision_lesson
(source=panel:deepseek+gemini); substance מדולג (INV-LRN5); הפיך + גיבוי CSV.
פאנל-הלכות: docstring/SCRIPTS.md עודכנו (--apply מחווט).
Frontend (web-ui/): כפתור "העלאת החלטה סופית של היו"ר" + שני כפתורים מדורגים
"הרץ למידת-קול"/"הרץ אימות-הלכות" ב-drafts-panel; כל התוויות בעברית
(badge מקור-לקח: "פאנל: דיפסיק+גמיני", "הרמס (סקירה)"...).
Spec: docs/spec/07-learning.md §0.6. Invariants: INV-LRN1/LRN4/LRN5, G10
(שער-יו"ר ידני להטמעה ל-SKILL.md/lessons.md — הפאנלים יוצרים הצעות בלבד);
G2 (מסלול-סופי הוא יכולת חסרה, לא מסלול-מקביל).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The 211 open missing_precedents include 99 Supreme serial-format rulings
(בג"ץ/בר"מ/עע"מ NNNN/YY) with no נט-format triple — fetchable only from
supremedecisions.court.gov.il. Decoded its public JSON API (no browser, no
CAPTCHA, no smart-card); validated live on בג"ץ 3483/05 + בר"מ 10212/16.
- court_fetch_supreme.py: rewrite. POST Home/SearchVerdicts with a structured
`document` ({Year:"YYYY", CaseNum, OldMainNumFormat:true, SearchText:[…]}) +
X-Requested-With header → records; GET Home/Download?path=&fileName=&type=4 →
PDF. The earlier attempt failed only on the request shape (string vs object).
2-digit→4-digit year; try candidate docs best-first (פסק-דין→pages), skipping
the published-report 's'-prefix files the free endpoint WAF-blocks.
- orchestrator: on successful ingest, close matching open missing_precedents
(link to the new case_law). End-to-end validated (בר"מ 10212/16 → corpus).
- backfill_missing_precedents.py: enqueue fetchable open gaps (supreme + net)
into court_fetch_jobs; the drainer fetches+ingests+closes. dry-run default.
- X13 spec + SCRIPTS.md updated (Tier-0 decoded, no longer a limitation).
Very old un-digitized Supreme cases (e.g. בג"ץ 389/87 → 0 records) → manual.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds docs/spec/X14-storage-minio.md — the domain spec + phased plan for
migrating binary document storage from the local data/ tree to the
already-deployed MinIO service (Coolify svc `minio`).
Captures: disk inventory, scattered file-I/O map (~8 services, no central
layer), DB path columns, MinIO deploy state, Paperclip = API-consumer only.
Defines 7 domain invariants (INV-STG1..7) and a 7-phase execution plan.
Chair decisions (2026-06-08): git-per-case keeps text/metadata + MinIO holds
binaries (INV-STG7); WORM Object-Lock on FINAL decisions only (INV-STG4);
internal Docker network for legal-ai↔MinIO.
Invariants: keeps G2 (single storage path replaces scattered I/O);
INV-STG1..7 new. Spec-only PR — no code/behavior change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Records the now-complete corpus citation graph: why native not Obsidian (G2),
the 6 opt-in node layers (precedent/topic/practice-area · halacha · gaps ·
digests), node size/color semantics, the Graph Analysis metrics
(PageRank/betweenness/community via web/graph_metrics.py), navigation, the
/api/graph/* endpoints, the key files, a how-to-extend recipe, the invariants
(G2/G5/UI2/UI4), and the PR history.
Adds docs/corpus-graph.md + a reference-table row in legal-ai/CLAUDE.md.
Docs only — no code change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The /precedents metadata queue was stuck — 24 rows requested, nothing draining
them — and the agentic claude CLI hit error_max_turns on what is a single
structured text→JSON task (slow + flaky). Metadata extraction is bounded
extraction, the wrong fit for an agentic loop.
- gemini_session.py: query_json drop-in (gemini-2.5-flash, JSON mode, httpx —
no new SDK dep). Reads GEMINI_API_KEY (~/.env; SoT Infisical
nautilus:/external-apis/gemini). Host-side only — no LLM from the container.
- precedent_metadata_extractor: claude_session.query_json → gemini_session.
Validated live: rich, accurate fields (case_name/summary/appeal_subtype/tags).
- process_pending_extractions: kind-aware cooldown — metadata 2s (Gemini, fast),
halacha keeps 30s (Claude rate limits).
- drain_metadata_queue.py + legal-metadata-drain.config.cjs (pm2 cron */15) so
the queue never clogs again. SCRIPTS.md.
- X8 INV-FP5 updated: per-task engine choice (Gemini=bounded metadata,
claude_session=agentic halacha), both host-side, single canonical queue (G2).
Agentic/voice-sensitive work (writing, analysis, halacha) stays on claude_session
(Daphna's subscription). Gemini cost ≈ $0.10/1M tokens — negligible.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The spec said "supreme → Tier-0"; reality (PR #124) routes by נט-format
availability — נט המשפט (Tier-1) serves all courts incl. Supreme-with-נט-format,
and only serial-only Supreme falls to the (still-unbuilt) Tier-0 → manual.
Updated §0 source-distinction, §1 routing diagram, §5 risks (Tier-0 limitation
+ scheduled drain). Docs-only.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
