Three new voice docs based on deep reading of 1033-25 (full-acceptance) and
7 representative cases for block-zayin (claims summary):
- daphna-acceptance-architecture.md: 5 distinct templates for case acceptance
(A: internal flaw + voiding; B: remand to committee; C: corrections in
request; D: substantive 8xxx; E: appraiser remand). Fixes the wrong
reference in architecture-by-outcome that treated full-acceptance as a
variation of partial-acceptance.
- daphna-block-zayin-claims.md: rules for claims summary block — order by
procedural role, neutrality, sub-headings per party, anti-patterns
(numbered lists, evaluation words, premature conclusion).
- daphna-decision-tree.md: operational tool that unifies all 5 voice docs
into a short analytical process. Starts with the decisive question:
"what is the winning evidence?". Decision trees for architecture
selection, opening mode, citation choice, length by weight.
Updates legal-writer.md to read decision-tree first, then the 5 voice docs,
plus block-zayin.md before block ז.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After reading all 23 1xxx decisions from style_corpus DB (in addition to
the 10 training files and 1130-25/1194-25 deep reads), synthesized two
new operational documents:
docs/daphna-precedent-network.md
- Maps each legal issue to the specific precedent Daphna cites
- 9 threshold issues (standing, השפר, סעיף 152, קנייני, פגמי פרסום,
פסילה, עבירות בנייה) with her preferred quotes for each
- 8 substantive issues (תכנון נקודתי vs כולל, חיקוק תכנית, סטייה ניכרת,
62א, חניה, תמ"א 38, תכניות ישנות, שימוש חורג)
- Lists ~30 external precedents she cites consistently + ~15 personal
precedents (her own canon — 1110/20 בעלז, 1112/22 שקופה, 1181/22 אדלר,
1130-25, etc.)
- Distinguishes precedents she cites vs. those she does NOT cite
docs/daphna-architecture-by-outcome.md
- 7 distinct block-yod architectures keyed to outcome type:
1. Pure rejection (short, 555-2000 words)
2. Rejection after complex analysis (2500-4500)
3. Threshold dismissal + merits "ועל מנת לא לצאת בחסר" (mode F)
4. Three or more distinct issues (sub-headings)
5. Partial acceptance (full funnel architecture)
6. Joined appeals
7. Remand follow-up
- Decision tree for the agent (4 questions → architecture choice)
- Internal proportions table (opening 5-10%, doctrine 15-25%, etc.)
- Costs matrix with 6 scenarios
Updated docs/daphna-voice-fingerprint.md with section 6 (additions from
23-file corpus read): 2 new opening modes (F: threshold+merits, G:
remand follow-up), nuanced sub-heading rule, self-citation of full
analytical blocks, 10 new "we" verbs, 11 traditional phrases with
sources, expanded costs matrix, transparency about petition outcomes,
warning that 1015-24 is dissent (not Daphna's voice).
Updated .claude/agents/legal-writer.md to require reading all 4 voice
docs before block-yod (the "voice quartet"), with explicit decision
tree integration.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After analyzing all 24 building_permit decisions in style_corpus DB
(not just the 2 local files), refined two anti-patterns:
1. Sub-headings: actually permitted when block-yod handles 3+ distinct
legal issues (e.g., 1079-24 had "הבקשות לפסילה" / "מעמד המבקשת
וזכות עמידה" / "עותרים ציבוריים"). The earlier rule of "no
sub-headings except academic cases" was too strict — based only on
small local sample.
2. Paragraph numbering: discovered it's an evolutionary pattern, not
a static rule. Pre-2025 decisions had sequential paragraph numbers
(1, 2, 3 throughout); recent decisions (1126-25, 1128-25, 1130-25,
1194-25) abandoned it for narrative flow. The agent should NOT add
paragraph numbers — the new style.
The (1)...(2)...(3)... in-paragraph enumeration ban remains absolute —
0/33 final decisions used it. Distinction now made explicit:
in-paragraph enumeration ≠ paragraph-level numbering (former always
forbidden; latter is evolutionary).
Updated:
- docs/daphna-voice-fingerprint.md — corpus stats, refined anti-patterns
- .claude/agents/legal-writer.md — checklist with new distinctions
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Synthesized two voice documents from corpus reading:
- docs/voice-1130-25.md: deep read of case 1130-25 block-yod (5000 words),
extracting the 9-movement funnel architecture, 8 reasoning templates,
10 'we' verbs with their distinct functions, the 'akhen...ulam' pattern,
pacing/silence principles, and the deliberative meta-narrative.
- docs/daphna-voice-fingerprint.md: cross-corpus synthesis of 10 finals
(1 planning + 9 appraisal levy). Identifies 10 invariants, 5 opening
modes mapped to outcome certainty, mandatory ברמ 3644/13 preamble for
shamai cases, copy-paste templates, and 7 anti-patterns to avoid.
Updated .claude/agents/legal-writer.md:
- Added voice docs as MUST-READ before block-yod (was missing the deep
voice layer; only had surface style_guide patterns)
- Replaced the ' (1)...(2)...(3)...' enumeration template with the 5 opening
modes (the enumeration was a known anti-pattern Daphna always removes)
- Added the 'we' verbs catalog with explicit functions
- Made 'אכן...אולם' pattern mandatory for issues with substantial
counter-arguments (was vaguely 'אמנם...אולם')
- Added mandatory ברמ 3644/13 preamble for 8xxx shamai cases
- Added self-citation triple-mode (refer/defer/distinguish) — Daphna's
emerging practice of building personal jurisprudence
- Added 8-item anti-patterns checklist for post-write review
- Replaced block-yod-alef section with proper 4-paragraph closing
template (process narrative → outcome → costs → date)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Paperclip auto-blocks any in_progress issue without a live execution path
within ~1 minute of the run finishing. When the CEO ends a run with an
@chaim question pending, the main case issue was staying in_progress and
getting auto-blocked, flooding the case timeline with "automatically
retried continuation" system comments (7 occurrences on 2026-04-16).
Add an explicit status protocol to the CEO instructions:
- in_review at the end of any run that leaves a pending @chaim question
- in_progress when resuming from user_commented (also at start of comment routing)
- done only after final export
Applied at all three @chaim waiting points (stages B/C) and at the top
of comment routing.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Fixes critical bug in 1033-25: user-uploaded עריכה-*.docx files were
orphaned on disk while exports kept rebuilding from stale DB blocks.
New architecture:
- User-uploaded DOCX becomes the source of truth (cases.active_draft_path)
- System edits via XML surgery with real Word <w:ins>/<w:del> revisions
- User can Accept/Reject each change from within Word
Components:
- docx_reviser.py: XML surgery for Track Changes (15 tests)
- docx_retrofit.py: retroactive bookmark injection with Hebrew marker
detection + heading heuristic (9 tests)
- docx_exporter.py: emits bookmarks around each of the 12 blocks
- 3 new MCP tools: apply_user_edit, list_bookmarks, revise_draft
- 4 new/updated endpoints: upload (auto-registers active draft),
/exports/revise, /exports/bookmarks, /exports/{filename}/retrofit,
/active-draft
- DB migration: cases.active_draft_path column
- UI: correct banner using real v-numbers, "מקור האמת" badge,
detailed upload toast with bookmarks_added/missing_blocks
- agents: legal-exporter (3 export modes), legal-ceo (stage G for
revision handling), legal-writer (revision mode)
Multi-tenancy:
- Works for both CMP (1xxx cases) and CMPA (8xxx/9xxx cases)
- New revise-draft skill added to both companies
- deploy-track-changes.sh syncs skills CMP ↔ CMPA
- retrofit_case.py: one-off retrofit of existing files
Tests: 34 passing (15 reviser + 9 retrofit + 4 exporter bookmarks + 6 e2e)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- legal-analyst: opus 4.6 → opus 4.7
- legal-proofreader: opus 4.6 → opus 4.7
- legal-writer: sonnet 4.6 → opus 4.7 (complex block writing benefits from stronger model)
- block_writer MODEL_MAP: updated opus ID to 4.7
Opus 4.7 brings: high-res images (2576px), better file-based memory,
improved DOCX generation, and task budgets for agentic loops.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New issues created by the CEO via curl were missing plugin_state records,
causing them to be invisible in the legal-ai UI. Added iron rule: after every
POST to create an issue, INSERT into plugin_state with the case number.
Also fixed 8070-25 CMPA issues directly in DB (3 records added).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- CEO agent now sends email via notify.py when awaiting human response
- CEO creates child issues (parentId) instead of flat disconnected issues
- Fix notify.py email address to chaim+paperclip@marcus-law.co.il
- Move Paperclip UI assets (RTL CSS + Hebrew JS) into repo under scripts/
- Add deploy.sh script to push assets to live Paperclip instance
- Fix comment box positioning: newest comment on top, input below it
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The CEO was ignoring the focused wake reason and doing a full heartbeat
scan of all cases/issues before getting to the actual comment. Added
step 0: check $PAPERCLIP_WAKE_REASON first — if user_commented, skip
directly to comment handling. Don't scan other cases.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
CEO was missing get_chair_directions, record_chair_feedback,
list_chair_feedback, and search_case_documents. Without these tools
it couldn't read or update chair directions when processing draft
annotations.
Now the CEO will:
1. Read existing chair_directions via MCP tool
2. Record each draft annotation as chair_feedback
3. Update analysis-and-research.md
4. Post summary for user review before routing to writer
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
When the user writes editing instructions inside a draft DOCX, the CEO
must not just forward them as a checklist. Instead:
1. Read analysis-and-research.md + existing chair_directions
2. Translate draft annotations into methodological structure (syllogism)
3. Update chair_directions with the new analysis
4. Post summary to user and WAIT for approval
5. Only after approval → create issue for writer
This gives the user a chance to verify the CEO understood correctly
before the writer starts working.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Even when the user asks to edit specific paragraphs in an existing
draft, the CEO must first analyze through the methodology: identify
which legal issue the edit serves, build syllogistic structure,
reference specific source documents, and state the review standard.
Without this, the writer gets a technical checklist instead of
methodological guidance.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Agent JWT cannot wake other agents directly (returns "Agent can only
invoke itself"). The correct pattern: create an issue + assign to the
target agent → Paperclip triggers wakeup automatically.
Also documented all correct API routes in HEARTBEAT.md:
- POST /api/issues/{id}/comments (not /issues/)
- POST /api/companies/{company-id}/issues (not /api/issues)
- PATCH /api/issues/{id}
- POST /api/agents/{id}/wakeup (self only, with payload.issueId)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The CEO was sending empty issues like "הועבר לכתיבה" without any
methodological content. The writer needs: syllogistic structure per
issue, source document references, claim handling table, chair
directions, style guidelines, and draft file path when available.
Added "תבנית issue לכותב ההחלטה" with all 5 required sections.
Updated comment routing to read drafts word-by-word and use the template.
Co-Authored-By: Claude Opus 4.6 (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>
When a user comments on a Paperclip issue, the built-in automation wakes
the assigned agent directly, bypassing the CEO. This meant user instructions
(like "read the uploaded draft and route to the right agent") were ignored.
Changes:
- Plugin: add issue.comment.created event handler that wakes the CEO agent
with the comment context (plugin-legal-ai, separate repo)
- HEARTBEAT: add steps 2b (read recent user comments) and 2c (check
attachments) before agents start working
- CEO agent: add comment-routing section — read, check attachments, route
- Writer agent: add step 0 — check for uploaded DOCX drafts before writing
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>
After Dafna fills her positions in the analysis document, the analyst
now runs a second pass to: verify cited case law against corpus and
case documents, deepen factual findings based on the chosen direction,
close open questions, and strengthen CREAC preparation.
Pipeline flow updated: direction_approved → analyst pass 2 →
analysis_enriched → CEO creates writer issue → ready_for_writing.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The status map was using informal descriptions ("מסמכים הוגהו")
instead of actual DB values. Now each row shows:
- The exact status string in cases.status
- Which agent sets it
- What the CEO should do next
New statuses added: proofread, analyst_verified, research_complete,
qa_passed, qa_failed, exported. Removed ambiguous conditions.
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>
Documents >15K chars must be split by chapter/section and extracted
in parts. If extract_claims times out, retry with chunks or extract
manually. This prevents the Matmon document issue (108K chars, 4x timeout).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Test run on case 1130-25 revealed critical gaps. This commit fixes:
HEARTBEAT.md (#1, #11):
- Agents MUST wake CEO after completing any task (wakeup request)
- New "blocked" status option — agents cannot mark "done" if something failed
- Fallback: direct DB insert if API wake doesn't work
legal-analyst.md (#2):
- New step 6: completeness checks BEFORE finishing
- Verify all appeal/response documents extracted successfully
- Verify all extracted documents produced claims
- Verify classification is correct (no claims from committee)
- If any check fails → status = "blocked", not "done"
legal-ceo.md (#3, #6, #7, #12, #13, #14, #15):
- Step A rewritten with 3 sub-checks:
A1: extraction completeness (no missing documents)
A2: negative checks (wrong classification, abnormal counts, missing parties)
A3: methodology compliance (syllogisms, CREAC prep, steel-man, etc.)
- Any failure blocks progress to step B
legal-qa.md (#6 reinforcement):
- New step 2b: negative checks on the written decision
- Missing issues, bare quotes, empty formulas, mixed findings/conclusions
Also:
- Synced all agent files to /home/chaim/legal-ai/ (Paperclip reads from there)
- Synced methodology + lessons + corpus docs
- Fixed claim classification in DB: 20 committee/applicant claims → response (#5)
Remaining gaps (3):
- #4: Paperclip cache may need restart to pick up new definitions
- #7: Matmon document retry (25K words, 0 claims extracted)
- #9: 53 appellant claims may need synthesis (high but not blocking)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
"בית ספר להחלטות" Phase 2 — the system now has formal analytical
methodology for building quasi-judicial decisions, separate from
Dafna's writing style (SKILL.md) and content checklists.
What was done:
- Downloaded 5 authoritative sources (~341K words): FJC Judicial
Writing Manual (1991+2020), Garner Legal Writing in Plain English,
Posner How Judges Think, Scalia/Garner Making Your Case
- Extracted principles from all sources into intermediate docs
- Synthesized into docs/decision-methodology.md (3,400 words,
12 sections, 10 guiding principles)
- Integrated methodology into block-yod prompt via {methodology_guidance}
- Restructured legal-writer agent workflow to follow analytical stages
- Made "answer all claims" flexible (bundle/skip via chair_directions)
- Added methodology compliance check (#7) to legal-qa agent
- Updated all knowledge files (CLAUDE.md, SKILL.md, lessons, corpus)
Three-layer architecture:
1. Methodology (decision-methodology.md) — universal, how to think
2. Content checklists (lessons.py) — specific per appeal subtype
3. Style (SKILL.md) — Dafna's personal writing patterns
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Closes the loop so דפנה's positions (written inline in the UI and
saved to analysis-and-research.md) automatically become binding
direction for the legal-writer agent — no manual copy-paste,
no bypass.
Backend:
- research_md.extract_chair_directions(path) returns a compact dict
with status (missing/empty/partial/complete), filled_count,
empty_count, and a reduced list of threshold_claims + issues each
with {id, number, title, direction}. Designed to be directly usable
as direction_doc by the writer.
- New MCP tool: drafting.get_chair_directions(case_number) wraps the
helper, resolves the case research file path via config.find_case_dir,
returns formatted JSON.
- Registered in server.py as mcp__legal-ai__get_chair_directions.
legal-writer agent update:
- Adds get_chair_directions to the tools list.
- New mandatory "שלב 1ב" before any block writing: call
get_chair_directions, branch on status.
- missing → halt, report "legal-analyst לא רץ עדיין"
- empty → halt, instruct Dafna to fill positions via the UI URL
- partial → halt unless user confirms; write only filled sections
- complete → proceed
- New "שלב 1ג" constructs an internal direction_doc from the
received chair rulings before writing block י.
- Block י section expanded with 5 binding rules:
1. Open each discussion with Dafna's ruling as the thesis
2. Frame the reasoning in her style (use get_style_guide phrases)
3. Match her tone (decisive vs nuanced)
4. Must NOT contradict her position — if she disagreed with your
own inclination, her position rules
5. Use legal_questions from the analysis file as the analytical
structure (principle question first, concrete application second)
- New bullet section for block יא: summarize each chair ruling
briefly, state final outcome, close with the signed date formula.
Verified all four status paths (missing/empty/partial/complete) via
local test. Now Dafna's workflow is fully end-to-end: she reads the
analyst report in the UI, fills "עמדת ועדת הערר" in each card, hits
blur to auto-save, then triggers legal-writer — which picks up her
positions as direction without any file shuffle.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add "עמדת ועדת הערר" placeholder to legal-analyst agent template
for each legal issue, to be filled by the chair as guidance for the
writing agent
- Fix green checkmark not showing for proofread documents (treat
'proofread' status same as 'completed')
- Show time alongside date in local files listing
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add delete_document_chunks for reprocessing, save extracted text to disk
- Expand case directory structure (original/extracted/proofread/backup)
- Update classifier patterns (תגובה, הודעת עמדה)
- Fix proofreader agent paths for new directory layout
- Update HEARTBEAT to notify on every task completion
- Improve bidi_table with LRE/PDF directional embedding
- Add Paperclip project verification and auto-close setup issue
- Add auto-sync-cases.sh for Gitea synchronization
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove cases/new|in-progress|completed subdivision (status managed in DB)
- Rename documents/original → documents/originals (consistent plural)
- Move exports from global data/exports/ into cases/{num}/exports/
- Add documents/research/ for case law and analysis files
- Update all agents, scripts, config, web API endpoints, and DB paths
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New: scripts/notify.py — sends via SMTP (notify@marcus-law.co.il → paperclip+chaim@marcus-law.co.il)
Updated: HEARTBEAT.md — agents must send email when waiting for human decision
Triggers: outcome choice, direction approval, QA failures, review ready.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
All agent output — comments, status, errors, summaries, thinking — must be in Hebrew.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Two issues that caused QA agent to fail:
1. save_block_content saved to DB only — now also rebuilds drafts/decision.md
2. legal-writer.md now has explicit mandatory step: case_update(status="drafted")
Without these, workflow_status reports has_draft=false and QA can't run.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
brainstorm_directions tool uses claude -p subprocess which times out
when called from inside a claude session (agent). CEO should think
about directions directly — it already has all the context.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
CEO now follows a step-by-step interactive flow:
A. Check status and what's been done
B. Summarize case + ask Chaim for outcome (1/2/3)
C. Read response, run brainstorm, present directions
D. Read direction choice, approve, launch writer agent
E. Monitor writing progress
F. QA and export
All interaction happens through Paperclip comments.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Manages the decision writing pipeline:
- Creates issues and assigns to specialist agents
- Tracks status across all active cases
- Reports to human (Chaim) when approvals needed
- Never writes or analyzes directly — delegates
All 4 specialist agents now report to CEO.
Hierarchy: עוזר משפטי → מנתח/חוקר/כותב/בודק
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Symlinked to Paperclip instructions directory for each agent.
Single source of truth: .claude/agents/ files → symlinked to Paperclip.
Cleaned duplicate soul_md from DB metadata.
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>
Defines the agent's role, tools, document type rules, and workflow.
Linked to Paperclip agent via --agent legal-analyst extraArg.
Key rules:
- Claims only from appeal docs, responses from response docs, replies from supplementary
- Never extract from precedents, plans, or protocols
- Must report results to Paperclip before finishing
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Ezer Mishpati - AI legal decision drafting system with:
- MCP server (FastMCP) with document processing pipeline
- Web upload interface (FastAPI) for file upload and classification
- pgvector-based semantic search
- Hebrew legal document chunking and embedding
