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>
חיווט ספ-המערכת לסוכני-Paperclip כך שכל סוכן חייב לקרוא את 00-constitution
תחילה, ואז את ספ-התחום הרלוונטי לתפקידו (לפי טבלת X4 §2) — לפני עבודה מהותית.
- HEARTBEAT.md: סעיף עליון "קריאת-ספ — קודם החוקה (00), אז ספ-התחום" לפני §0–§8,
עם טבלת תפקיד→ספ ל-8 הסוכנים.
- 8 קבצי-סוכן (ceo/proofreader/researcher/analyst/writer/qa/exporter/hermes):
סעיף "קרא לפני פעולה (INV-AG1)" בראש הגוף.
- X4-agents.md: שדה "אכיפה" של INV-AG1 → "מחוּוט (פרוצדורלי)"; §5 → "בוצע".
אכיפה פרוצדורלית בכוונה — invariant פרויקטלי-תפעולי, אין שער-קוד שמכריח קריאה.
prereq לסוכני-התהליך (תת-פרויקט 5). gap-audit נשמר כ-snapshot (כמו FU-8a).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Root cause of "agent can't find the Agasi decision in the corpus" (CMPA-55):
the decision was fully ingested, but the retrieval layer failed on the
realistic agent query — searching by case name.
- RC-A (#52): lexical tsvector covered only chunk content + halacha text,
so a bare-name query ("אגסי") matched decisions that *cite* the case, not
the case itself. Add meta_tsv on case_law(case_name, case_number) (SCHEMA
V20) and OR it into the lexical halacha/chunk SQL with a match boost, so a
name/number hit surfaces the case's own rows. Agasi: rank 4 → rank 1.
- RC-B (#53): precedent_library_list hard-defaulted source_kind=external_upload
and never exposed the param, hiding uploaded ערר/בל"מ (internal_committee)
decisions. Thread source_kind through service → tool → MCP tool (supports
'internal_committee' / 'all_committees').
- #54: agent instructions (researcher/analyst/writer) — search-by-name
protocol: add content/case-number, search both corpora, use all_committees
before declaring "not in corpus".
- #55: chunker produced tiny fragment chunks ("דיון", "החלטה") from header
keywords matched mid-sentence. Anchor SECTION_PATTERNS to line start +
merge sub-min sections; exclude <50-char fragments at query time (484
existing fragments hidden; full re-chunk tracked as #57).
Tests: scripts/test_retrieval_by_name.py (name ranks case above citer +
substantive regressions); chunker unit checks (0 tiny chunks). New findings
filed as tasks #56 (halacha source_kind leak) and #57 (re-chunk migration).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Same case_number can exist as both a regular appeal (ערר) and an
extension-of-time request (בל"מ), and we were inferring the difference
from appeal_subtype prefixes — fragile, and case-number lookups
weren't disambiguated. Now stored as a first-class field on both
case_law (corpus) and cases (live cases), with partial unique indexes
on (case_number, proceeding_type).
- SCHEMA_V15: column + CHECK constraints + backfill from
appeal_subtype LIKE 'extension_request_%' + partial unique indexes
replace the old global UNIQUE(case_number).
- derive_proceeding_type() centralizes the inference rule
(extension_request_* → בל"מ; subject regex fallback; default ערר).
- Metadata extractor prompt asks Claude to populate the new field
explicitly; apply_to_record writes it for internal_committee rows.
- internal_decision_upload, case_create, case_update accept an
optional proceeding_type; FastAPI request models expose it.
- Wizard + edit dialog get a sided Select; case header renders the
resolved label (ערר / בל"מ).
- Uploaded the 2 staged בל"מ decisions on betterment levy:
8126/24 (סופר נוח, 13 chunks), 8047/23 (הרנון, 48 chunks).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Four parallel sub-agents closed the remaining critical gaps from the
26/05 Stage A/B sprint. Each block independently tested; aggregated here.
## #30/#31 finalizers (sub-agent A)
* Auto-derive practice_area in case_create from case_number prefix
(1xxx→rishuy_uvniya, 8xxx→betterment_levy, 9xxx→compensation_197);
default for CaseCreateRequest is now "" (the DB constraint catches
any stray "appeals_committee").
* practice_area.py: derive_subtype now handles axis-B domain values
(rishuy_uvniya/betterment_levy/compensation_197) without parsing the
case number; new helper derive_domain_practice_area().
* Halacha re-extraction verified unnecessary — all 6 reclassified
records already had is_binding=false and approved halachot.
* Regression tests: 6 cases in tests/test_corpus_constraints.py
covering practice_area enum, internal-committee chair/district,
external-upload arar prefix, MCP guard.
* UI: district input → Select dropdown (7 districts) in
precedent-edit-sheet.tsx, preserving legacy free-text values.
## #37 בל"מ subtypes (sub-agent B)
* 3 new appeal_subtypes: extension_request_{building_permit,
betterment_levy,compensation}. APPEALS_COMMITTEE_SUBTYPES extended,
SUBTYPES_BY_AREA mappings added.
* New helpers: is_blam_subject(), is_blam_subtype(),
derive_subtype_with_blam(case_number, subject, practice_area).
case_create now uses it to auto-detect "בקשה להארכת מועד" subjects.
* 3 methodology templates under docs/methodology/extension-request-*.md.
* paperclip_client.py mapping updated for the 3 new subtypes
(extension_request_building_permit→CMP, the other two→CMPA).
* Frontend: bilingual "בל"מ" badge + filter dropdown on cases list +
detail header; appeal-type-bars collapseBlam() merges בל"מ into its
parent domain for aggregate bars.
* Wizard auto-detects בל"מ from subject during case creation.
* 3 Berlinger cases (1017/1018/1019-03-26) migrated to
appeal_subtype=extension_request_building_permit via psql.
## #35 missing_precedents feature (sub-agent C)
* Schema V13: missing_precedents table (citation, case_id, party,
legal_topic, status, linked_case_law_id, claim_quote, ...) +
FK constraints + 3 indexes. Applied via psql + idempotent migration.
* 6 db.py service functions, 3 MCP tools, 6 FastAPI endpoints
(POST/GET/PATCH/DELETE/upload — upload routes by citation prefix
to ingest_internal_decision or ingest_precedent).
* Next.js page /missing-precedents with 5 status tabs + filters +
sidebar badge counter + detail drawer with metadata edit + smart
upload form that switches fields per committee/court.
* Bootstrap: 7 rows imported from the JSON file
(3 citations × cases, all status=closed with linked_case_law_id).
* legal-researcher.md: new §2ב.5 with missing_precedent_create
usage + dedup semantics + tool grant.
## #36 legal_arguments aggregation (sub-agent D)
* Schema V14: legal_arguments + legal_argument_propositions M:M.
Applied via psql.
* New service argument_aggregator.py with two functions —
aggregate_claims_to_arguments() (Claude CLI / claude_session) and
get_legal_arguments(). Graceful llm_unavailable handling when CLI
is missing (containers).
* 2 MCP tools + 2 API endpoints (POST .../aggregate-arguments as
BackgroundTask, GET .../legal-arguments).
* Frontend: shadcn Accordion + new legal-arguments-panel.tsx with
hierarchical (party → priority badge → arguments) display, "טיעונים"
tab on the case page, "חשב/חשב מחדש" buttons.
* scripts/backfill_legal_arguments.py + SCRIPTS.md entry — dry-run
found 8 candidate cases including 1017/1018/1019.
## Open follow-ups (intentionally deferred)
* npm run api:types in web-ui (CLAUDE.md flow) — recommended before
the next UI commit; not required for backend deployment.
* Run backfill_legal_arguments.py --apply once the container picks up
the new aggregator service.
* webhook on missing-precedents upload-close to Paperclip (optional).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Model drift (instructions → match DB):
- CEO: claude-sonnet-4-6 → claude-opus-4-6 (DB runs opus; CEO needs opus quality)
- מנתח/כותב/מגיה: claude-opus-4-7 → claude-opus-4-6 (DB runs 4-6; no 4-7 in adapter)
legal-proofreader.md:
- {issue-id} placeholder → $PAPERCLIP_TASK_ID בשני המקומות (done + blocked)
legal-researcher.md:
- הוסף reference ל-HEARTBEAT.md בראש הקובץ
legal-qa.md:
- הבהרת שיטת בדיקת corpus_queries_logged: grep ידני בלבד, לא validate_decision
CLAUDE.md (curator):
- הוסף תהליך אישור הצעות curator: comment → חיים מאשר → commits ל-SKILL.md/lessons.md
maxConcurrentRuns CEO: כבר 2 ב-DB — לא נדרש שינוי
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
CEO (legal-ceo.md):
- הסרת company UUID ו-project UUID קשוחים בדוגמת יצירת issue
- שימוש ב-$PAPERCLIP_COMPANY_ID לחברה
- project_id נשלף דינמית מה-issue ההורה דרך $PAPERCLIP_TASK_ID
researcher (legal-researcher.md):
- הוסף mcp__legal-ai__search_internal_decisions לרשימת tools
- הוסף סעיף 2ב.2א המסביר את ההבדל: search_decisions = דפנה בלבד;
search_internal_decisions = כל ועדות הערר בכל המחוזות
- הוראות מתי להשתמש + אזהרת היררכיה (ועדת ערר < מחוזי)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Brings the legal-ai ↔ Paperclip integration in line with the official
Paperclip skill. Net effect: HEARTBEAT.md -47% (370→195 lines), all 14
agents on uniform runtime_config + budget + instructionsBundleMode, and
two cross-company helpers replacing manual SQL.
Highlights:
- HEARTBEAT.md refactor: project-specific only, delegates to the official
paperclipai/paperclip skill (loaded per agent). Adds heartbeat-context
fast-path (§1.7) and PAPERCLIP_WAKE_PAYLOAD_JSON shortcut (§1.5).
- Issue Thread Interactions API: legal-ceo.md now uses
ask_user_questions / request_confirmation / suggest_tasks instead of
free-text comments — gives chair structured UI with idempotency keys.
- pc.sh + paperclip_api.pc_request: every API call goes through helpers
that inject Authorization + X-Paperclip-Run-Id (audit trail).
- sync_agents_across_companies.py: master(CMP)→mirror(CMPA) sync via
Paperclip API, idempotent, with --verify and --apply modes.
- skills/new-company-setup: 11-step blueprint distilling all 11 gaps
into a single onboarding runbook for the next company.
- .taskmaster: 12 tasks covering each gap (one already closed: #29).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When Paperclip wakes the CEO and the model issues an mcp__legal-ai__*
call within ~10s of session init, Claude Code sometimes returns
"No such tool available" because the legal-ai MCP server hasn't
finished bringing up its tool catalog yet. Observed twice today on
CMPA precedent-extraction wakeups (sessions 9989fbaf and a9c61801);
the agent fell back to bash + .venv/bin/python and finished the work,
but the race needed fixing on the server side.
Three changes that close the window:
1. Lazy schema init (services/db.py + server.py)
`init_schema()` was awaited inside the FastMCP lifespan, blocking
the `initialize`/`tools/list` handshake until ~10 CREATE TABLE IF
NOT EXISTS statements ran. Under contention (two CEOs waking at
once for different companies) this stretched. Now the lifespan
returns immediately and `get_pool()` runs the schema migrations
exactly once on first DB access, guarded by an asyncio.Lock.
tools/list is answered in milliseconds regardless of DB state.
2. Lazy heavy imports
- services/embeddings.py: voyageai (~450ms) loaded only inside
_get_client()
- services/extractor.py: google.cloud.vision (~550ms) loaded only
inside _get_vision_client() and _ocr_with_google_vision()
These two were being imported at module top from
legal_mcp.tools.documents -> services.processor -> services.{
extractor,embeddings}, so the FastMCP server couldn't even start
responding until both finished. Cold start dropped from 2.7s to
1.17s end-to-end (init + tools/list response).
3. Agent-side warmup + retry guidance (.claude/agents/legal-ceo.md)
Even with a fast server, the model can still race on the very
first call. The precedent-extraction section now tells the CEO
to call workflow_status as a warmup probe and to retry after a
short sleep if it sees "No such tool available", before falling
back to the python bypass.
Also expanded the precedent-tool whitelists on the sub-agents that
delegate halacha/library work (commits 4a9a6b7 + 7ee90dc added the
tools to the MCP server but only the CEO got them in its allowed
list). Added to: legal-researcher (full extraction set), legal-analyst
(library_get/list + halacha review), legal-writer (library lookups +
halacha_review), legal-qa (library_get + halacha_review), and the two
that the CEO was already missing (halacha_review, halachot_pending).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a third corpus of legal authority distinct from style_corpus
(Daphna's prior decisions for voice) and case_precedents (chair-attached
quotes per case). The new corpus holds chair-uploaded court rulings and
other appeals committee decisions, with binding rules (הלכות) extracted
automatically and queued for chair approval.
Pipeline (web/app.py + services/precedent_library.py):
file → extract → chunk → Voyage embed → halacha_extractor → store +
publish progress over the existing Redis SSE channel.
Schema V7 (services/db.py): extends case_law with source_kind +
extraction status fields under a CHECK constraint pinning practice_area
to the three appeals committee domains (rishuy_uvniya, betterment_levy,
compensation_197). New precedent_chunks (vector(1024)) and halachot
tables (vector(1024) over rule_statement, IVFFlat indexes, gin on
practice_areas/subject_tags). Halachot start as pending_review; only
approved/published rows are visible to search_precedent_library.
Agents: legal-writer, legal-researcher, legal-analyst, legal-ceo,
legal-qa get search_precedent_library. legal-writer prompt explains
the three-corpus distinction and CREAC use; legal-qa now verifies that
every cited halacha resolves to an approved row in the corpus.
UI: /precedents page with four tabs — library / semantic search /
pending review (J/K nav, A/R/E shortcuts, badge count) / stats.
Reuses the existing upload-sheet progress + SSE pattern.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two structural gaps in legal-researcher's "שלב 5: דיווח" surfaced while
auditing the case 8174-24 run:
1. **No DB linkage.** The skill told the researcher to post a comment
summarizing precedents but never to call mcp__legal-ai__precedent_attach.
The MCP tool itself wasn't even in the tools frontmatter — so even
a researcher that wanted to write to case_precedents physically
couldn't. Result: 0 rows in case_precedents after a successful
research run, even with 8 precedents identified and verified in
the comment text. The writer then has to grep free-text instead
of querying a structured table.
2. **No persisted file.** Research output existed only as a Paperclip
comment. The writer/QA can't `Read` it from disk; they have to go
through Paperclip API to fetch comment bodies. Compare to the
analyst, which is required to write `analysis-and-research.md`.
Fix:
• Added precedent_attach, precedent_list, precedent_search_library
to the tools frontmatter.
• Rewrote step 5 with explicit ordering: save to disk → attach
verified precedents to DB → update status → email → post comment.
• Documented the precedent_attach call signature inline (case_number,
citation, quote, section_id) so the agent doesn't have to reverse-
engineer it. Includes guidance on which precedents to attach
(verified with quote) vs which to leave for external verification.
Effect: future research runs will populate case_precedents and
data/cases/{N}/documents/research/precedent-research.md, both of which
the writer needs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The retry loop bug we fixed in legal-analyst yesterday existed in every
single sub-agent skill. They all post a comment + wake the CEO + exit,
leaving their own issue in `in_progress`. Paperclip's "in_progress with
no live execution" watchdog then re-wakes them, repeating until something
external transitions the issue. Watched it happen on CMPA-17 (researcher)
today — 4 iterations + manual SIGTERM + manual PATCH.
Same fix applied to all 5 remaining agents:
• legal-researcher.md
• legal-writer.md
• legal-qa.md
• legal-exporter.md
• legal-proofreader.md (file was incomplete — also added the missing
שלב 5: דיווח and wake-CEO sections to bring it to parity with the
other agents)
Each gets a "סגור את ה-issue של עצמך — חובה!" section with two PATCH
templates: one for `done` after a successful run, one for `blocked` if
checks fail or output is incomplete. The section sits before the
wake-CEO block, with an explicit reference to the CMPA-17 incident so
the rule has a concrete anchor.
Result: every agent now has the same close-issue contract. No more
zombie in_progress issues, no more 4× wakeup loops.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Until now only legal-writer referenced the voice corpus. Without these
references the qa agent can't validate writer output, the researcher
chooses precedents outside Daphna's canon, and the analyst's claims
classification doesn't match block-zayin rules.
- legal-qa: adds 8th check "voice_compliance" — block ז structure,
block י voice (אכן/אולם, "אנחנו" verbs, no numbered lists), correct
precedent from canon, acceptance template match.
- legal-researcher: must check daphna-precedent-network.md before
proposing any precedent; cross-reference with Daphna's own past
decisions via search_decisions.
- legal-analyst: reads block-zayin-claims.md — its output is the
writer's input for block ז.
- legal-ceo: lists all 6 voice docs and which agent reads each.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The agents used /api/agents/{id}/wake (404) with a fallback of INSERT
INTO agent_wakeup_requests. The DB insert creates only the wakeup
record without a heartbeat_run, so the Paperclip dispatcher never
processes it — agents get stuck in queued forever.
Fix:
- All agents: /wake → /wakeup (correct Paperclip API endpoint)
- Remove all DB INSERT fallbacks, replace with warning
- Document the rule in CLAUDE.md: always API, never DB insert
- Save to memory for future conversations
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Install git in Docker image and wrap all subprocess git calls in
try/except so a missing or failing git binary never kills an upload
that already succeeded.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Every agent now has explicit instructions in its own definition file,
not just in HEARTBEAT.md. An agent following only its own step-by-step
instructions will do the right thing on any new case.
All 6 non-CEO agents: explicit wakeup CEO block in completion step
(curl API + psql fallback, with agent name customized)
legal-ceo.md: issue template for analyst with 5 mandatory items
(document mapping table, no-extract list, split large docs,
wakeup CEO, blocked if failed)
legal-writer.md: explicit Read of decision-methodology.md as step 1
(before case_get, not just "read before starting")
legal-qa.md: methodology_compliance severity → critical
(was warning — decisions without syllogisms/steel-man now blocked)
legal-proofreader.md: added case_update tool + status='proofread'
(was missing entirely — CEO couldn't know proofreading was done)
legal-researcher.md: added case_update + mail notification
(was missing — CEO couldn't know research was done)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Complete agent pipeline for decision writing:
1. legal-analyst (existing) — extract claims/responses/replies
2. legal-researcher (new) — analyze precedents, plans, protocols
3. legal-writer (new) — write decision blocks in Dafna's style
4. legal-qa (new) — validate before export (6 checks)
All agents use claude_local adapter (Claude Code session, zero API cost).
Each has YAML frontmatter with specific tools and detailed Hebrew instructions.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
