fix: add debug log and null company_id comment to webhook scheduling

asyncpg returns JSONB columns as raw JSON strings when no type codec is
configured (only pgvector is registered in _init_connection). The stored
value is a correct JSONB array (jsonb_typeof=array confirmed), but
asyncpg decodes it as str. Parse it explicitly in the GET handler so
the frontend receives the correct Python list/dict.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

.claude/agents/hermes-curator.md

@@ -0,0 +1,146 @@
+---
+name: hermes-curator
+description: Knowledge Curator (Hermes) — מנתח החלטות סופיות אחרי export, מציע עדכונים ל-skills/lessons. read-only על תוכן, write רק על comments.
+adapter: deepseek_local
+model: deepseek-v4-pro
+profiles:
+  CMP:  curator-cmp   # רישוי ובניה (תיקים 1xxx)
+  CMPA: curator-cmpa  # היטל השבחה + פיצויים (תיקים 8xxx, 9xxx)
+---
+
+> **Why DeepSeek**: A/B test 2026-05-05 הראה ש-DeepSeek V4-Pro חזק יותר מ-Sonnet
+> על דפוסי סגנון/לקסיקון, פי 2-3 מהיר, פי ~20 זול. הסוכן לא דורש דייקנות עובדתית
+> על תוצאת התיק (זו עבודתו של ה-CEO/Writer/QA), לכן הטיה מקרית של DeepSeek בקריאת
+> תוצאה לא משפיעה על איכות הסקירה.
+
+# מנהל ידע — Hermes Knowledge Curator
+
+## רקע
+
+אני סוכן Hermes Agent (לא Claude Code), מותקן בתור POC לבדיקה האם Hermes
+מתאים יותר מ-Claude Code לתפקידי ניתוח עם זיכרון ארוך-טווח.
+
+קיימים שני מופעים שלי — אחד לכל חברה — עם profile וזיכרון נפרדים:
+- **CMP** (תיקים 1xxx): רישוי ובניה. profile=`curator-cmp`. UUID `60dce831-...`
+- **CMPA** (תיקים 8xxx + 9xxx): היטלי השבחה ופיצויים. profile=`curator-cmpa`. UUID `d6f7c55d-...`
+
+**איך אני מופעל:** דפנה לוחצת "סמן כסופי" בקובץ ב-UI של legal-ai →
+`POST /api/cases/{case_number}/exports/{filename}/mark-final` רץ ב-`web/app.py` →
+הוא קורא ל-`pc_wake_curator_for_final()` ב-`web/paperclip_client.py` שיוצר
+לי sub-issue ומעיר אותי. **לא דרך CEO** — חיבור ישיר מהאירוע ב-UI לסוכן.
+זה מבטיח שאני מנתח את הגרסה האמיתית של דפנה, לא טיוטה אינטרמדיאטית.
+
+ה-CEO (`עוזר משפטי`, `claude_local`) ממשיך להיות ה-orchestrator של כל
+התהליך עד שלב F (ייצוא DOCX) ו-G (טיפול בעריכות). אני לא מחליף אותו —
+מוסיף שכבת ניתוח אחרי שדפנה החליטה שהגרסה הסופית מוכנה.
+
+**אינטראקציה במקום comments חופשיים:** ה-promptTemplate שלי תומך ב-3 סוגי
+`issue_thread_interactions` של Paperclip. כשאני מסיים ניתוח, אני בוחר אחד
+לפי הקונטקסט:
+
+- `ask_user_questions` — multi-select של ממצאים שדפנה תרצה לקדם ל-style guide
+- `request_confirmation` — אישור/דחייה לפעולה ספציפית (עם detailsMarkdown מורחב)
+- `suggest_tasks` — הצעת issues חדשים לפעולה (Paperclip יוצר אותם אם דפנה אישרה)
+
+ה-UI של legal-ai מציג אותם דרך `agent-activity-feed.tsx` (commit `d099470`):
+רדיו / checkbox / accept-reject buttons. דפנה עונה — Paperclip מעיר אותי
+שוב עם `$PAPERCLIP_APPROVAL_ID`, ואני מעבד את התשובה ב-§B של ה-promptTemplate.
+
+## תפקיד
+
+לאחר שכל החלטה סופית מיוצאת ל-DOCX, אני נקרא לסקור אותה. המטרה:
+לזהות **דפוסים חדשים** או **פערים** שיכולים לשפר את ה-style guide
+ואת ה-lessons לעתיד.
+
+יו"ר הוועדה היא עו"ד דפנה תמיר. **אני לא מחליף את שיקול דעתה** — רק
+מציע נקודות שיכולות להיות שימושיות לעדכון מסמכי ייחוס.
+
+## מה אני עושה בכל wake
+
+1. קורא את ה-issue body שב-`{{taskBody}}` — שם התיק + ID של ההחלטה הסופית
+2. משתמש ב-MCP tools של legal-ai:
+   - `mcp__legal-ai__case_get` — קבלת פרטי תיק (כולל `expected_outcome` — **הסמכות העובדתית** לתוצאה)
+   - `mcp__legal-ai__case_get_final_text` — הטקסט המלא של ההחלטה הסופית
+   - `mcp__legal-ai__document_list` — רק אם נדרש רשימת מסמכים נוספים של התיק
+   - `mcp__legal-ai__get_style_guide` — דפוסי הסגנון של דפנה
+   - **לא** להשתמש ב-`search_decisions` — השוואה ל-`SKILL.md` ו-`corpus-analysis.md` מספיקה ולא יקרה
+3. קורא קבצים מקומיים (read-only):
+   - `/home/chaim/legal-ai/skills/decision/SKILL.md`
+   - `/home/chaim/legal-ai/docs/legal-decision-lessons.md`
+   - `/home/chaim/legal-ai/docs/corpus-analysis.md`
+4. מעדכן את `~/.hermes/profiles/curator-cmp/memories/MEMORY.md` עם ממצאים
+   (Hermes שומר אוטומטית — אני יכול גם להשתמש ב-memory tool)
+5. כותב comment על ה-issue הזה דרך Paperclip API:
+   ```
+   POST {{paperclipApiUrl}}/issues/{{taskId}}/comments
+   Authorization: Bearer $PAPERCLIP_API_KEY
+   { "body": "<my findings>" }
+   ```
+6. סוגר את ה-issue (status=done) אחרי שכתבתי את ה-comment
+
+## פורמט ה-comment
+
+עברית, ניטרלי. 3-5 ממצאים מובחנים. **כל ממצא חייב להיות מתויג** באחד מ-4 הסוגים:
+
+```
+[סגנון]          — מילים, ביטויי מעבר, פתיחות, סיומים
+[מבנה]           — סדר בלוקים, יחסי אורך, מספור
+[לקסיקון משפטי] — מינוח טכני (מגישי תכנית, ריפוי פגם, וכו')
+[טבלאי]          — דפוסים שמופיעים פעמיים+ ב-corpus
+```
+
+לכל ממצא:
+- **מה ראיתי** — תיאור קצר של הדפוס/הפער
+- **מה זה אומר** — למה זה חשוב
+- **הצעה** — איך אפשר להוסיף ל-style guide / lessons (טקסט מוצע מילולי)
+
+אם אין ממצאים חדשים → לציין במפורש בלי להמציא.
+
+## מה **לא** להגיד ב-comment
+
+- **אל תכלול שורת מטא** בראש ה-comment עם "תוצאה: X" או "אורך: ~Y תווים".
+  אתה לא בודק את התיק — אתה בודק את הסגנון. תוצאה מוטעית בראש ה-comment פוגעת באמינות.
+- אם תוצאה רלוונטית להמחשת דפוס מסוים — קח אותה **מ-`case_get` (`expected_outcome`)**, **לא מקריאת הטקסט**.
+  אם השדה ריק או חסר ב-DB — סמן `[תוצאה: לא מאומתת]` או דלג עליה.
+- **אל תפרש משפטית** את ההחלטה. דפנה כבר הכריעה. תפקידך זיהוי דפוסים בלבד.
+
+## מה אני לא עושה
+
+- **לא מעדכן** קבצים בעצמי (skills/, lessons.py, DB) — רק מציע
+- **לא יוצר** issues חדשים
+- **לא מעיר** סוכנים אחרים
+- **לא דן** עם המשתמש על תוכן ההחלטה — רק מנתח דפוסים
+
+## כשאני נכשל
+
+אם MCP server לא נגיש או החלטה לא נמצאת, כתוב comment קצר עם הסיבה
+ו-status=failed. אל תזייף ממצאים.
+
+## דרישות מ-`deepseek_local` adapter (חובה)
+
+ה-adapter שמריץ אותי **חייב** להזריק 3 דברים בכל wake — אחרת interactions ייחסמו ב-`401 "Agent run id required"`:
+
+1. **env `PAPERCLIP_API_KEY`** — agent's own pcp_ key
+2. **env `PAPERCLIP_RUN_ID`** — ה-`heartbeat_runs.id` של ה-wake הנוכחי
+3. **env `PAPERCLIP_API_URL`** + **`PAPERCLIP_TASK_ID`** — לקריאות API
+
+ב-`hermes_local` (`adapters/registry.ts:240-288`) ההזרקה הזו נעשית אוטומטית, ובנוסף Paperclip prepends auth-guard לפני ה-promptTemplate. ב-`deepseek_local` החדש — לוודא שמיושם.
+
+ה-promptTemplate **כבר** כולל את ה-header `X-Paperclip-Run-Id: $PAPERCLIP_RUN_ID` בכל קריאת mutating (POST/PATCH), כך שאם ה-adapter רק מזריק את ה-env vars נכון, ה-interactions יעבדו ישירות בלי תלות ב-auth-guard injection.
+
+### Verification:
+
+```bash
+# על תיק חי, אחרי שדפנה לוחצת mark-final, ה-curator יקבל:
+echo "PAPERCLIP_RUN_ID=$PAPERCLIP_RUN_ID"  # חייב להיות UUID חוקי
+echo "PAPERCLIP_API_KEY=${PAPERCLIP_API_KEY:0:8}..."  # חייב להתחיל ב-pcp_
+echo "PAPERCLIP_API_URL=$PAPERCLIP_API_URL"  # חייב להיות http://localhost:3100/api
+```
+
+## קונטקסט קבוע (לא לשכוח)
+
+- היו"ר: עו"ד דפנה תמיר
+- חברה: ועדת ערר רישוי ובניה (CMP, תיקים 1xxx)
+- שפה: עברית בלבד
+- 24 החלטות במאגר האימון, 12-block architecture, סגנון דפנה
+- אני קורא מ-MEMORY.md בכל wake — שם הקונטקסט שלי מצטבר

.claude/agents/legal-ceo.md

@@ -87,6 +87,7 @@ tools:
 | כותב החלטה | 7ed8686f-24bc-49a3-bc02-67ca15b895a9 | כתיבת בלוקים ה-יב (Opus) |
 | בודק איכות | 1a5b229e-9220-4b13-940c-f8eb7285fc29 | QA לפני ייצוא |
 | מייצא טיוטה | d0dc703b-ca83-4883-bca7-c9449e8713cd | בדיקה סופית + ייצוא DOCX מגורסת |
+| מנהל ידע (Hermes) | CMP: 60dce831-5c5b-4bae-bda9-5282d506f0dc · CMPA: d6f7c55d-570a-46b8-8d72-1286d07da0d8 | סקירת החלטות סופיות, הצעות לעדכון style guide / lessons. **לא קורא ישירות מ-CEO** — מופעל אוטומטית מ-`web/app.py:api_mark_final` כשדפנה לוחצת "סמן כסופי" ב-UI. |
 
 ## כלל: כל issue חדש = תת-משימה
 

.taskmaster/tasks/tasks.json

@@ -1362,21 +1362,22 @@
         "description": "מסך אדמין ב-legal-ai UI שמציג את כל הגדרות הסוכנים (model, timeout, runtime_config, skills, budget) ומאפשר עריכה ושמירה. מונע SQL ישיר.",
         "details": "מטרה: ממשק אדמין מרכזי במקום שעריכה תהיה רק ב-UI של Paperclip + SQL ישיר + CUSTOMIZATIONS.md.\n\nשדות (לכל סוכן × 2 חברות):\n1. adapter_config: model, effort, timeoutSec, maxTurnsPerRun, extraArgs[], paperclipSkillSync.desiredSkills[]\n2. runtime_config.heartbeat: graceSec, cooldownSec, wakeOnDemand, maxConcurrentRuns, enabled, intervalSec\n3. budget_monthly_cents (לקראת gap #2)\n4. status / pause_reason (קריאה + כפתור pause/resume)\n\nאופציות מימוש:\nA. עמוד חדש ב-legal-ai/web-ui (Next.js 16) — קורא Paperclip DB דרך FastAPI endpoint חדש (/api/admin/paperclip-agents)\nB. קריאה ל-Paperclip API (/api/companies/{id}/agents) — REST טהור, פחות שדות זמינים\nC. iframe ל-Paperclip UI — שטחי\n\nהמלצה: A. שולט מלא + ולידציה משפטית (timeoutSec >= 1800 כי OCR).\n\nתלוי ב: gap #25 (סוכנים מוכפלים) — אם נעבור לshared, המסך יתאים.\n\nשאלות פתוחות לחיים:\n- auth: מי יכול לגשת? (כיום אין auth ב-legal-ai)\n- bulk edit ל-2 חברות יחד או נפרד?\n- חשיפת skill marketplace (להוסיף/להוריד skills) או רק קריאה?",
         "testStrategy": "1. עמוד עולה ב-/admin/agents בlegal-ai UI. 2. עריכת timeoutSec ושמירה → SELECT ב-DB מאמת. 3. revision חדש ב-agent_config_revisions עם source מתאים.",
-        "status": "pending",
+        "status": "done",
         "dependencies": [
           "16",
           "17",
           "25"
         ],
         "priority": "medium",
-        "subtasks": []
+        "subtasks": [],
+        "updatedAt": "2026-05-04T17:29:25.686Z"
       }
     ],
     "metadata": {
       "version": "1.0.0",
-      "lastModified": "2026-05-04T16:44:27.554Z",
+      "lastModified": "2026-05-04T17:29:25.687Z",
       "taskCount": 29,
-      "completedCount": 23,
+      "completedCount": 24,
       "tags": [
         "legal-ai"
       ]

CLAUDE.md

@@ -118,6 +118,8 @@
 ├── web-ui/                            ← Next.js frontend (TypeScript/React): ממשק המשתמש
 │   └── next.config.ts                    ← proxy: /api/* → FastAPI :8000
 ├── mcp-server/                        ← MCP server + services + tools
+├── adapters/                          ← Paperclip external adapters (ראה למטה)
+│   └── deepseek-paperclip-adapter/      ← `deepseek_local` (Hermes-pinned ל-DeepSeek profile)
 └── scripts/                           ← סקריפטים וכלי עזר (ראה scripts/SCRIPTS.md)
     └── .archive/                      ← סקריפטים שהושלמו (לא להריץ)
 ```
@@ -180,6 +182,15 @@
 - הסקריפט מסנן local skills שלא קיימים ב-CMPA (מציג אזהרה), משתמש ב-API (לא DB ישיר), יוצר revisions, idempotent.
 - שאלות ה-skill הרשמי של Paperclip — `paperclip` skill תחת `paperclipai/paperclip`.
 
+### External adapters — `deepseek_local`
+- מיקום ה-package: [adapters/deepseek-paperclip-adapter/](adapters/deepseek-paperclip-adapter/) (לא ב-`node_modules`).
+- רישום ב-Paperclip: רשומה ב-`~/.paperclip/adapter-plugins.json` (נטען אוטומטית ב-startup דרך `buildExternalAdapters`). אין צורך בעריכת `node_modules`.
+- **מה ה-adapter עושה**: spawnל-`hermes chat` עם `HERMES_HOME=/home/chaim/.hermes/profiles/deepseek` כך שה-CLI טוען את `config.yaml` (`base_url=https://api.deepseek.com/v1`, `provider=custom`, `key_env=DEEPSEEK_API_KEY`) ואת `.env` (שמכיל את ה-key).
+- **מודלים זמינים** (lookup ב-DeepSeek `/v1/models`): `deepseek-v4-pro` (default), `deepseek-v4-flash`. יופיעו כדרופ-דאון ב-UI.
+- **התקנה מחדש / עדכון**: `curl -X POST -H "Authorization: Bearer pcapi_legal_install_key_2026" -H "Content-Type: application/json" -d '{"packageName":"/home/chaim/legal-ai/adapters/deepseek-paperclip-adapter","isLocalPath":true}' http://localhost:3100/api/adapters/install`. לעדכון hot — `POST /api/adapters/deepseek_local/reload`.
+- **⚠ Cross-company sync**: `sync_agents_across_companies.py` **מדלג** על סוכנים עם `adapter_type` שונה בין CMP ל-CMPA. כשעוברים סוכן ל-`deepseek_local` חובה להחיל ידנית בשתי החברות לפני sync.
+- **תוספת adapters עתידיים** (OpenAI ישיר, Anthropic ישיר, וכו'): אותו דפוס. ה-package הראשי חייב לייצא `createServerAdapter()` שמחזיר `{ type, label, models, agentConfigurationDoc, execute, testEnvironment, sessionCodec, listSkills, syncSkills, ... }`. ראה את [adapters/deepseek-paperclip-adapter/dist/index.js](adapters/deepseek-paperclip-adapter/dist/index.js) כתבנית.
+
 ---
 
 ## עקרונות כתיבה קריטיים

adapters/deepseek-paperclip-adapter/dist/index.js

@@ -0,0 +1,99 @@
+/**
+ * DeepSeek (via Hermes) — external Paperclip adapter.
+ *
+ * Loaded by Paperclip's plugin-loader. Contract:
+ *   The package's main module must export createServerAdapter() returning
+ *   a single ServerAdapterModule object with all fields wired in.
+ *
+ * Runtime: spawns the local `hermes` CLI with HERMES_HOME pinned to a
+ * DeepSeek profile that defines model.base_url=https://api.deepseek.com/v1
+ * and model.key_env=DEEPSEEK_API_KEY.
+ */
+
+import {
+  ADAPTER_TYPE,
+  ADAPTER_LABEL,
+  DEEPSEEK_MODELS,
+  DEFAULT_PROFILE_HOME,
+} from "./shared/constants.js";
+import { execute } from "./server/execute.js";
+import { testEnvironment } from "./server/test.js";
+import { sessionCodec } from "./server/session-codec.js";
+import { listSkills, syncSkills } from "./server/skills.js";
+
+const AGENT_CONFIGURATION_DOC = `# DeepSeek (via Hermes) — Agent Configuration
+
+DeepSeek-pinned variant of the Hermes adapter. Runs the local \`hermes\` CLI
+with \`HERMES_HOME\` pointed at a DeepSeek profile (\`config.yaml\` declares
+\`base_url=https://api.deepseek.com/v1\` and \`key_env=DEEPSEEK_API_KEY\`).
+
+## Prerequisites
+
+- Hermes Agent installed (\`pip install hermes-agent\`) — \`hermes --version\` works.
+- DeepSeek profile dir exists (default: \`/home/chaim/.hermes/profiles/deepseek\`)
+  with \`config.yaml\` + \`.env\` (containing \`DEEPSEEK_API_KEY\`).
+
+## Core Configuration
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| model | string | \`deepseek-v4-pro\` | DeepSeek model id (\`deepseek-v4-pro\` or \`deepseek-v4-flash\`). |
+| provider | string | \`custom\` | Hermes provider name. The DeepSeek profile defines \`provider: custom\` so \`custom\` is the right value. |
+| hermesProfileHome | string | \`/home/chaim/.hermes/profiles/deepseek\` | Absolute path to a Hermes profile dir. Set per-agent if you maintain multiple DeepSeek profiles. |
+| timeoutSec | number | 1800 | Execution timeout in seconds. |
+| graceSec | number | 30 | SIGTERM grace period in seconds. |
+
+## Tools / Workspace
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| toolsets | string | (profile default) | Comma-separated toolsets to enable. |
+| persistSession | boolean | true | Resume sessions across heartbeats via \`--resume\`. |
+| worktreeMode | boolean | false | Use git worktree for isolated changes. |
+| checkpoints | boolean | false | Enable filesystem checkpoints. |
+
+## Advanced
+
+| Field | Type | Default | Description |
+|-------|------|---------|-------------|
+| hermesCommand | string | \`hermes\` | Path to the hermes binary. |
+| verbose | boolean | false | Enable verbose Hermes logs. |
+| extraArgs | string[] | [] | Extra CLI args appended after standard flags. |
+| env | object | {} | Extra environment variables passed to Hermes. \`HERMES_HOME\` here overrides \`hermesProfileHome\`. |
+| promptTemplate | string | (default) | Override the default Paperclip wakeup prompt. |
+| paperclipApiUrl | string | \`http://127.0.0.1:3100/api\` | Paperclip API URL injected into the prompt template. |
+
+## Available template variables
+
+\`{{agentId}}\`, \`{{agentName}}\`, \`{{companyId}}\`, \`{{companyName}}\`,
+\`{{runId}}\`, \`{{taskId}}\`, \`{{taskTitle}}\`, \`{{taskBody}}\`,
+\`{{commentId}}\`, \`{{wakeReason}}\`, \`{{projectName}}\`, \`{{paperclipApiUrl}}\`.
+`;
+
+export function createServerAdapter() {
+  return {
+    type: ADAPTER_TYPE,
+    label: ADAPTER_LABEL,
+    models: DEEPSEEK_MODELS,
+    agentConfigurationDoc: AGENT_CONFIGURATION_DOC,
+
+    execute,
+    testEnvironment,
+    sessionCodec,
+    listSkills,
+    syncSkills,
+
+    // Capability flags
+    supportsLocalAgentJwt: true,
+    supportsInstructionsBundle: false,
+    requiresMaterializedRuntimeSkills: false,
+  };
+}
+
+// Also export the loose constants for any caller that wants to inspect
+// the package without invoking createServerAdapter (e.g., test harnesses).
+export const type = ADAPTER_TYPE;
+export const label = ADAPTER_LABEL;
+export const models = DEEPSEEK_MODELS;
+export const agentConfigurationDoc = AGENT_CONFIGURATION_DOC;
+export const defaultProfileHome = DEFAULT_PROFILE_HOME;

adapters/deepseek-paperclip-adapter/dist/server/execute.js

@@ -0,0 +1,352 @@
+/**
+ * Server-side execution for the DeepSeek-via-Hermes adapter.
+ *
+ * Spawns `hermes chat -q "..." -Q -m <model> --provider custom` with
+ * HERMES_HOME pinned to a DeepSeek-configured profile so the same machine
+ * can run other Hermes-based agents on different providers in parallel.
+ *
+ * The Hermes CLI loads model.base_url, model.key_env (DEEPSEEK_API_KEY),
+ * and toolsets from <HERMES_HOME>/config.yaml + <HERMES_HOME>/.env.
+ */
+
+import {
+  runChildProcess,
+  buildPaperclipEnv,
+  renderTemplate,
+  ensureAbsoluteDirectory,
+} from "@paperclipai/adapter-utils/server-utils";
+import {
+  HERMES_CLI,
+  DEFAULT_PROFILE_HOME,
+  DEFAULT_MODEL,
+  DEFAULT_PROVIDER,
+  DEFAULT_TIMEOUT_SEC,
+  DEFAULT_GRACE_SEC,
+  SESSION_ID_REGEX,
+  SESSION_ID_REGEX_LEGACY,
+  TOKEN_USAGE_REGEX,
+  COST_REGEX,
+} from "../shared/constants.js";
+
+function cfgString(v) {
+  return typeof v === "string" && v.length > 0 ? v : undefined;
+}
+function cfgNumber(v) {
+  return typeof v === "number" ? v : undefined;
+}
+function cfgBoolean(v) {
+  return typeof v === "boolean" ? v : undefined;
+}
+function cfgStringArray(v) {
+  return Array.isArray(v) && v.every((i) => typeof i === "string") ? v : undefined;
+}
+
+const DEFAULT_PROMPT_TEMPLATE = `You are "{{agentName}}", an AI agent employee in a Paperclip-managed company powered by DeepSeek.
+
+IMPORTANT: Use the \`terminal\` tool with \`curl\` for ALL Paperclip API calls (web_extract and browser cannot access localhost).
+
+Your Paperclip identity:
+  Agent ID: {{agentId}}
+  Company ID: {{companyId}}
+  API Base: {{paperclipApiUrl}}
+
+{{#taskId}}
+## Assigned Task
+
+Issue ID: {{taskId}}
+Title: {{taskTitle}}
+
+{{taskBody}}
+
+## Workflow
+
+1. Work on the task using your tools.
+2. When done, mark the issue completed:
+   \`curl -s -X PATCH "{{paperclipApiUrl}}/issues/{{taskId}}" -H "Content-Type: application/json" -d '{"status":"done"}'\`
+3. Post a completion comment summarizing what you did:
+   \`curl -s -X POST "{{paperclipApiUrl}}/issues/{{taskId}}/comments" -H "Content-Type: application/json" -d '{"body":"DONE: <your summary here>"}'\`
+{{/taskId}}
+
+{{#commentId}}
+## Comment on This Issue
+
+Someone commented. Read it:
+   \`curl -s "{{paperclipApiUrl}}/issues/{{taskId}}/comments/{{commentId}}" | python3 -m json.tool\`
+Address the comment, POST a reply if needed, then continue working.
+{{/commentId}}
+
+{{#noTask}}
+## Heartbeat Wake — Check for Work
+
+1. List your open issues:
+   \`curl -s "{{paperclipApiUrl}}/companies/{{companyId}}/issues?assigneeAgentId={{agentId}}"\`
+2. Pick the highest priority and work on it. When done, follow steps 2-3 above.
+3. If nothing to do, report briefly what you checked.
+{{/noTask}}`;
+
+function buildPrompt(ctx, config) {
+  const template = cfgString(config.promptTemplate) || DEFAULT_PROMPT_TEMPLATE;
+  const taskId = cfgString(ctx.context?.taskId);
+  const taskTitle = cfgString(ctx.context?.taskTitle) || "";
+  const taskBody = cfgString(ctx.context?.taskBody) || "";
+  const commentId = cfgString(ctx.context?.commentId) || "";
+  const wakeReason = cfgString(ctx.context?.wakeReason) || "";
+  const agentName = ctx.agent?.name || "DeepSeek Agent";
+  const companyName = cfgString(ctx.context?.companyName) || "";
+  const projectName = cfgString(ctx.context?.projectName) || "";
+
+  let paperclipApiUrl =
+    cfgString(config.paperclipApiUrl) ||
+    process.env.PAPERCLIP_API_URL ||
+    "http://127.0.0.1:3100/api";
+  if (!paperclipApiUrl.endsWith("/api")) {
+    paperclipApiUrl = paperclipApiUrl.replace(/\/+$/, "") + "/api";
+  }
+
+  const vars = {
+    agentId: ctx.agent?.id || "",
+    agentName,
+    companyId: ctx.agent?.companyId || "",
+    companyName,
+    runId: ctx.runId || "",
+    taskId: taskId || "",
+    taskTitle,
+    taskBody,
+    commentId,
+    wakeReason,
+    projectName,
+    paperclipApiUrl,
+  };
+
+  let rendered = template;
+  rendered = rendered.replace(/\{\{#taskId\}\}([\s\S]*?)\{\{\/taskId\}\}/g, taskId ? "$1" : "");
+  rendered = rendered.replace(/\{\{#noTask\}\}([\s\S]*?)\{\{\/noTask\}\}/g, taskId ? "" : "$1");
+  rendered = rendered.replace(/\{\{#commentId\}\}([\s\S]*?)\{\{\/commentId\}\}/g, commentId ? "$1" : "");
+  return renderTemplate(rendered, vars);
+}
+
+function cleanResponse(raw) {
+  return raw
+    .split("\n")
+    .filter((line) => {
+      const t = line.trim();
+      if (!t) return true;
+      if (t.startsWith("[tool]") || t.startsWith("[hermes]") || t.startsWith("[paperclip]") || t.startsWith("[deepseek]")) return false;
+      if (t.startsWith("session_id:")) return false;
+      if (/^\[\d{4}-\d{2}-\d{2}T/.test(t)) return false;
+      if (/^\[done\]\s*┊/.test(t)) return false;
+      if (/^┊\s*[\p{Emoji_Presentation}]/u.test(t) && !/^┊\s*💬/.test(t)) return false;
+      if (/^\p{Emoji_Presentation}\s*(Completed|Running|Error)?\s*$/u.test(t)) return false;
+      return true;
+    })
+    .map((line) => {
+      let t = line.replace(/^[\s]*┊\s*💬\s*/, "").trim();
+      t = t.replace(/^\[done\]\s*/, "").trim();
+      return t;
+    })
+    .join("\n")
+    .replace(/\n{3,}/g, "\n\n")
+    .trim();
+}
+
+function parseHermesOutput(stdout, stderr) {
+  const combined = stdout + "\n" + stderr;
+  const result = {};
+
+  const sessionMatch = stdout.match(SESSION_ID_REGEX);
+  if (sessionMatch?.[1]) {
+    result.sessionId = sessionMatch[1];
+    const sessionLineIdx = stdout.lastIndexOf("\nsession_id:");
+    if (sessionLineIdx > 0) {
+      result.response = cleanResponse(stdout.slice(0, sessionLineIdx));
+    }
+  } else {
+    const legacyMatch = combined.match(SESSION_ID_REGEX_LEGACY);
+    if (legacyMatch?.[1]) result.sessionId = legacyMatch[1];
+    const cleaned = cleanResponse(stdout);
+    if (cleaned.length > 0) result.response = cleaned;
+  }
+
+  const usageMatch = combined.match(TOKEN_USAGE_REGEX);
+  if (usageMatch) {
+    result.usage = {
+      inputTokens: parseInt(usageMatch[1], 10) || 0,
+      outputTokens: parseInt(usageMatch[2], 10) || 0,
+    };
+  }
+
+  const costMatch = combined.match(COST_REGEX);
+  if (costMatch?.[1]) result.costUsd = parseFloat(costMatch[1]);
+
+  if (stderr.trim()) {
+    const errorLines = stderr
+      .split("\n")
+      .filter((line) => /error|exception|traceback|failed/i.test(line))
+      .filter((line) => !/INFO|DEBUG|warn/i.test(line));
+    if (errorLines.length > 0) result.errorMessage = errorLines.slice(0, 5).join("\n");
+  }
+
+  return result;
+}
+
+export async function execute(ctx) {
+  const config = ctx.agent?.adapterConfig ?? {};
+
+  const hermesCmd = cfgString(config.hermesCommand) || HERMES_CLI;
+  const model = cfgString(config.model) || DEFAULT_MODEL;
+  const provider = cfgString(config.provider) || DEFAULT_PROVIDER;
+  const profileHome = cfgString(config.hermesProfileHome) || DEFAULT_PROFILE_HOME;
+  const timeoutSec = cfgNumber(config.timeoutSec) || DEFAULT_TIMEOUT_SEC;
+  const graceSec = cfgNumber(config.graceSec) || DEFAULT_GRACE_SEC;
+  const toolsets = cfgString(config.toolsets) || cfgStringArray(config.enabledToolsets)?.join(",");
+  const extraArgs = cfgStringArray(config.extraArgs);
+  const persistSession = cfgBoolean(config.persistSession) !== false;
+  const worktreeMode = cfgBoolean(config.worktreeMode) === true;
+  const checkpoints = cfgBoolean(config.checkpoints) === true;
+  const useQuiet = cfgBoolean(config.quiet) !== false;
+
+  const prompt = buildPrompt(ctx, config);
+
+  const args = ["chat", "-q", prompt];
+  if (useQuiet) args.push("-Q");
+  if (model) args.push("-m", model);
+  args.push("--provider", provider);
+  if (toolsets) args.push("-t", toolsets);
+  if (worktreeMode) args.push("-w");
+  if (checkpoints) args.push("--checkpoints");
+  if (cfgBoolean(config.verbose) === true) args.push("-v");
+  args.push("--source", "tool");
+  args.push("--yolo");
+
+  const prevSessionId = cfgString(ctx.runtime?.sessionParams?.sessionId);
+  if (persistSession && prevSessionId) args.push("--resume", prevSessionId);
+  if (extraArgs?.length) args.push(...extraArgs);
+
+  // Pin Hermes to the DeepSeek profile by default. The agent can override
+  // by setting adapter_config.hermesProfileHome or adapter_config.env.HERMES_HOME.
+  const env = {
+    ...process.env,
+    ...buildPaperclipEnv(ctx.agent),
+    HERMES_HOME: profileHome,
+  };
+  if (ctx.runId) env.PAPERCLIP_RUN_ID = ctx.runId;
+  const taskId = cfgString(ctx.context?.taskId);
+  if (taskId) env.PAPERCLIP_TASK_ID = taskId;
+
+  // Parity with hermes_local (paperclip-src/server/src/adapters/registry.ts:267):
+  // inject the per-run agent auth token so the agent can call the Paperclip API.
+  // Without this, every Paperclip API write from the running agent fails with 401.
+  //
+  // Resolve env from the runtime-resolved config (ctx.config.env contains plain
+  // strings — Paperclip's secrets service unwraps {type:"plain"|"secret_ref", ...}
+  // bindings before invocation in services/heartbeat.ts:5433-5437).
+  // Fall back to agent.adapterConfig.env with manual unwrapping for older paths.
+  function unwrapEnvValue(v) {
+    if (typeof v === "string") return v;
+    if (v && typeof v === "object" && !Array.isArray(v)) {
+      if (v.type === "plain" && typeof v.value === "string") return v.value;
+    }
+    return undefined; // skip secret_ref / unknown types — let resolver handle them
+  }
+  const resolvedUserEnv =
+    ctx.config && typeof ctx.config === "object" && ctx.config.env && typeof ctx.config.env === "object" && !Array.isArray(ctx.config.env)
+      ? ctx.config.env
+      : null;
+  const rawUserEnv =
+    typeof config.env === "object" && config.env !== null && !Array.isArray(config.env)
+      ? config.env
+      : {};
+  // Prefer pre-resolved values from ctx.config.env when available; fall back to
+  // unwrapping raw bindings from agent.adapterConfig.env.
+  const flattenedUserEnv = {};
+  for (const [k, v] of Object.entries(rawUserEnv)) {
+    const resolved = resolvedUserEnv && typeof resolvedUserEnv[k] === "string" ? resolvedUserEnv[k] : unwrapEnvValue(v);
+    if (typeof resolved === "string") flattenedUserEnv[k] = resolved;
+  }
+  const userEnvApiKey = flattenedUserEnv.PAPERCLIP_API_KEY;
+  const explicitApiKey =
+    typeof userEnvApiKey === "string" && userEnvApiKey.trim().length > 0;
+  if (ctx.authToken && !explicitApiKey) env.PAPERCLIP_API_KEY = ctx.authToken;
+
+  // Apply unwrapped user env (may override HERMES_HOME, OPENAI_API_KEY, etc.).
+  Object.assign(env, flattenedUserEnv);
+
+  const cwd = cfgString(config.cwd) || cfgString(ctx.config?.workspaceDir) || ".";
+  try {
+    await ensureAbsoluteDirectory(cwd);
+  } catch {
+    // non-fatal
+  }
+
+  await ctx.onLog(
+    "stdout",
+    `[deepseek] Starting Hermes (model=${model}, provider=${provider}, profileHome=${env.HERMES_HOME}, timeout=${timeoutSec}s)\n`,
+  );
+  if (prevSessionId) {
+    await ctx.onLog("stdout", `[deepseek] Resuming session: ${prevSessionId}\n`);
+  }
+
+  // Reclassify benign Hermes stderr lines as stdout so the UI doesn't paint them red.
+  const wrappedOnLog = async (stream, chunk) => {
+    if (stream === "stderr") {
+      const trimmed = chunk.trimEnd();
+      const isBenign =
+        /^\[?\d{4}[-/]\d{2}[-/]\d{2}T/.test(trimmed) ||
+        /^[A-Z]+:\s+(INFO|DEBUG|WARN|WARNING)\b/.test(trimmed) ||
+        /Successfully registered all tools/.test(trimmed) ||
+        /MCP [Ss]erver/.test(trimmed) ||
+        /tool registered successfully/.test(trimmed) ||
+        /Application initialized/.test(trimmed);
+      if (isBenign) return ctx.onLog("stdout", chunk);
+    }
+    return ctx.onLog(stream, chunk);
+  };
+
+  // Forward ctx.onSpawn so Paperclip persists processPid/processGroupId to the
+  // heartbeat_runs row. Without it, the reaper cannot verify the child is alive
+  // (run.processPid is null) and treats the run as orphaned during long quiet
+  // phases (DeepSeek V4-Pro thinking can be silent for 60-90s per turn).
+  const result = await runChildProcess(ctx.runId, hermesCmd, args, {
+    cwd,
+    env,
+    timeoutSec,
+    graceSec,
+    onLog: wrappedOnLog,
+    onSpawn: ctx.onSpawn,
+  });
+
+  const parsed = parseHermesOutput(result.stdout || "", result.stderr || "");
+  await ctx.onLog(
+    "stdout",
+    `[deepseek] Exit code: ${result.exitCode ?? "null"}, timed out: ${result.timedOut}\n`,
+  );
+  if (parsed.sessionId) {
+    await ctx.onLog("stdout", `[deepseek] Session: ${parsed.sessionId}\n`);
+  }
+
+  const executionResult = {
+    exitCode: result.exitCode,
+    signal: result.signal,
+    timedOut: result.timedOut,
+    provider,
+    model,
+  };
+  if (parsed.errorMessage) executionResult.errorMessage = parsed.errorMessage;
+  if (parsed.usage) executionResult.usage = parsed.usage;
+  if (parsed.costUsd !== undefined) executionResult.costUsd = parsed.costUsd;
+  if (parsed.response) executionResult.summary = parsed.response.slice(0, 2000);
+
+  executionResult.resultJson = {
+    result: parsed.response || "",
+    session_id: parsed.sessionId || null,
+    usage: parsed.usage || null,
+    cost_usd: parsed.costUsd ?? null,
+  };
+
+  if (persistSession && parsed.sessionId) {
+    executionResult.sessionParams = { sessionId: parsed.sessionId };
+    executionResult.sessionDisplayId = parsed.sessionId.slice(0, 16);
+  }
+
+  return executionResult;
+}

adapters/deepseek-paperclip-adapter/dist/server/session-codec.js

@@ -0,0 +1,29 @@
+/**
+ * Session codec — Hermes uses a single sessionId for cross-heartbeat continuity
+ * via the --resume CLI flag. Same shape as the Hermes adapter.
+ */
+
+function readNonEmptyString(value) {
+  return typeof value === "string" && value.trim().length > 0 ? value.trim() : null;
+}
+
+export const sessionCodec = {
+  deserialize(raw) {
+    if (typeof raw !== "object" || raw === null || Array.isArray(raw)) return null;
+    const sessionId =
+      readNonEmptyString(raw.sessionId) ?? readNonEmptyString(raw.session_id);
+    if (!sessionId) return null;
+    return { sessionId };
+  },
+  serialize(params) {
+    if (!params) return null;
+    const sessionId =
+      readNonEmptyString(params.sessionId) ?? readNonEmptyString(params.session_id);
+    if (!sessionId) return null;
+    return { sessionId };
+  },
+  getDisplayId(params) {
+    if (!params) return null;
+    return readNonEmptyString(params.sessionId) ?? readNonEmptyString(params.session_id);
+  },
+};

adapters/deepseek-paperclip-adapter/dist/server/skills.js

@@ -0,0 +1,171 @@
+/**
+ * Skill snapshot for the DeepSeek-via-Hermes adapter.
+ *
+ * Hermes manages its own skills under ~/.hermes/skills/ (global; not per-profile).
+ * Paperclip-managed skills declared in adapter config are surfaced as
+ * "company_managed" entries — same behavior as the upstream Hermes adapter.
+ */
+
+import fs from "node:fs/promises";
+import path from "node:path";
+import { fileURLToPath } from "node:url";
+import {
+  readPaperclipRuntimeSkillEntries,
+  resolvePaperclipDesiredSkillNames,
+} from "@paperclipai/adapter-utils/server-utils";
+import { ADAPTER_TYPE } from "../shared/constants.js";
+
+const __moduleDir = path.dirname(fileURLToPath(import.meta.url));
+
+function asString(value) {
+  return typeof value === "string" && value.trim().length > 0 ? value.trim() : null;
+}
+
+function parseSkillFrontmatter(content) {
+  const match = content.match(/^---\s*\n([\s\S]*?)\n---/);
+  if (!match) return {};
+  const fm = {};
+  for (const line of match[1].split("\n")) {
+    const idx = line.indexOf(":");
+    if (idx === -1) continue;
+    const key = line.slice(0, idx).trim();
+    let val = line.slice(idx + 1).trim();
+    if ((val.startsWith('"') && val.endsWith('"')) || (val.startsWith("'") && val.endsWith("'"))) {
+      val = val.slice(1, -1);
+    }
+    fm[key] = val;
+  }
+  return fm;
+}
+
+async function buildSkillEntry(key, skillMdPath, categoryPath) {
+  let description = null;
+  try {
+    const content = await fs.readFile(skillMdPath, "utf8");
+    description = parseSkillFrontmatter(content).description ?? null;
+  } catch {
+    // ignore
+  }
+  return {
+    key,
+    runtimeName: key,
+    desired: true,
+    managed: false,
+    state: "installed",
+    origin: "user_installed",
+    originLabel: "Hermes skill",
+    locationLabel: `~/.hermes/skills/${categoryPath}`,
+    readOnly: true,
+    sourcePath: skillMdPath,
+    targetPath: null,
+    detail: description,
+  };
+}
+
+async function scanHermesSkills(skillsHome) {
+  const entries = [];
+  try {
+    const cats = await fs.readdir(skillsHome, { withFileTypes: true });
+    for (const cat of cats) {
+      if (!cat.isDirectory()) continue;
+      const catPath = path.join(skillsHome, cat.name);
+      const topSkill = path.join(catPath, "SKILL.md");
+      if (await fs.stat(topSkill).catch(() => null)) {
+        entries.push(await buildSkillEntry(cat.name, topSkill, cat.name));
+      }
+      const items = await fs.readdir(catPath, { withFileTypes: true }).catch(() => []);
+      for (const item of items) {
+        if (!item.isDirectory()) continue;
+        const skillMd = path.join(catPath, item.name, "SKILL.md");
+        if (await fs.stat(skillMd).catch(() => null)) {
+          entries.push(await buildSkillEntry(item.name, skillMd, `${cat.name}/${item.name}`));
+        }
+      }
+    }
+  } catch {
+    // ~/.hermes/skills/ doesn't exist
+  }
+  return entries.sort((a, b) => a.key.localeCompare(b.key));
+}
+
+async function buildSnapshot(config) {
+  const homedir =
+    asString(config.env?.HOME) ??
+    process.env.HOME ??
+    "/home/chaim";
+  const hermesSkillsHome = path.join(homedir, ".hermes", "skills");
+
+  const paperclipEntries = await readPaperclipRuntimeSkillEntries(config, __moduleDir);
+  const desiredSkills = resolvePaperclipDesiredSkillNames(config, paperclipEntries);
+  const desiredSet = new Set(desiredSkills);
+  const availableByKey = new Map(paperclipEntries.map((e) => [e.key, e]));
+
+  const hermesSkillEntries = await scanHermesSkills(hermesSkillsHome);
+  const hermesKeys = new Set(hermesSkillEntries.map((e) => e.key));
+
+  const entries = [];
+  const warnings = [];
+
+  for (const entry of paperclipEntries) {
+    const desired = desiredSet.has(entry.key);
+    entries.push({
+      key: entry.key,
+      runtimeName: entry.runtimeName,
+      desired,
+      managed: true,
+      state: desired ? "configured" : "available",
+      origin: entry.required ? "paperclip_required" : "company_managed",
+      originLabel: entry.required ? "Required by Paperclip" : "Managed by Paperclip",
+      readOnly: false,
+      sourcePath: entry.source,
+      targetPath: null,
+      detail: desired ? "Will be available on the next run via Hermes skill loading." : null,
+      required: Boolean(entry.required),
+      requiredReason: entry.requiredReason ?? null,
+    });
+  }
+
+  for (const entry of hermesSkillEntries) {
+    if (availableByKey.has(entry.key)) continue;
+    entries.push(entry);
+  }
+
+  for (const desired of desiredSkills) {
+    if (availableByKey.has(desired) || hermesKeys.has(desired)) continue;
+    warnings.push(`Desired skill "${desired}" is not available in Paperclip or Hermes skills.`);
+    entries.push({
+      key: desired,
+      runtimeName: null,
+      desired: true,
+      managed: true,
+      state: "missing",
+      origin: "external_unknown",
+      originLabel: "External or unavailable",
+      readOnly: false,
+      sourcePath: null,
+      targetPath: null,
+      detail: "Cannot find this skill in Paperclip or ~/.hermes/skills/.",
+    });
+  }
+
+  return {
+    adapterType: ADAPTER_TYPE,
+    supported: true,
+    mode: "persistent",
+    desiredSkills,
+    entries,
+    warnings,
+  };
+}
+
+export async function listSkills(ctx) {
+  return buildSnapshot(ctx.config);
+}
+
+export async function syncSkills(ctx, _desired) {
+  return buildSnapshot(ctx.config);
+}
+
+export function resolveDesiredSkillNames(config, availableEntries) {
+  return resolvePaperclipDesiredSkillNames(config, availableEntries);
+}

adapters/deepseek-paperclip-adapter/dist/server/test.js

@@ -0,0 +1,164 @@
+/**
+ * Environment test for the DeepSeek (via Hermes) adapter.
+ */
+
+import { execFile } from "node:child_process";
+import { promisify } from "node:util";
+import fs from "node:fs/promises";
+import path from "node:path";
+import {
+  HERMES_CLI,
+  ADAPTER_TYPE,
+  DEFAULT_PROFILE_HOME,
+} from "../shared/constants.js";
+
+const execFileAsync = promisify(execFile);
+
+function asString(v) {
+  return typeof v === "string" ? v : undefined;
+}
+
+async function checkCliInstalled(command) {
+  try {
+    await execFileAsync(command, ["--version"], { timeout: 10_000 });
+    return null;
+  } catch (err) {
+    if (err && err.code === "ENOENT") {
+      return {
+        level: "error",
+        message: `Hermes CLI "${command}" not found in PATH`,
+        hint: "Install Hermes Agent: pip install hermes-agent",
+        code: "deepseek_hermes_cli_not_found",
+      };
+    }
+    return null;
+  }
+}
+
+async function checkProfile(profileHome) {
+  try {
+    const stat = await fs.stat(profileHome);
+    if (!stat.isDirectory()) {
+      return {
+        level: "error",
+        message: `Profile path is not a directory: ${profileHome}`,
+        hint: "Create the directory or override hermesProfileHome in adapter config.",
+        code: "deepseek_profile_not_dir",
+      };
+    }
+  } catch {
+    return {
+      level: "error",
+      message: `Hermes profile dir does not exist: ${profileHome}`,
+      hint: "Create the profile dir with config.yaml + .env (DEEPSEEK_API_KEY).",
+      code: "deepseek_profile_missing",
+    };
+  }
+
+  const configPath = path.join(profileHome, "config.yaml");
+  try {
+    await fs.stat(configPath);
+  } catch {
+    return {
+      level: "error",
+      message: `Profile is missing config.yaml: ${configPath}`,
+      hint: "Add config.yaml with model.default + model.base_url + model.key_env.",
+      code: "deepseek_profile_no_config",
+    };
+  }
+
+  return {
+    level: "info",
+    message: `Profile resolved: ${profileHome}`,
+    code: "deepseek_profile_ok",
+  };
+}
+
+async function checkApiKey(profileHome, configEnv) {
+  // 1. config.env (resolved by Paperclip from secrets)
+  if (configEnv && typeof configEnv === "object" && asString(configEnv.DEEPSEEK_API_KEY)) {
+    return {
+      level: "info",
+      message: "DEEPSEEK_API_KEY found in adapter env config",
+      code: "deepseek_api_key_in_config",
+    };
+  }
+  // 2. Profile-local .env
+  try {
+    const envFile = path.join(profileHome, ".env");
+    const text = await fs.readFile(envFile, "utf-8");
+    if (/^\s*DEEPSEEK_API_KEY=/m.test(text)) {
+      return {
+        level: "info",
+        message: `DEEPSEEK_API_KEY found in ${envFile}`,
+        code: "deepseek_api_key_in_profile",
+      };
+    }
+  } catch {
+    // ignore
+  }
+  // 3. Process env
+  if (process.env.DEEPSEEK_API_KEY) {
+    return {
+      level: "info",
+      message: "DEEPSEEK_API_KEY found in Paperclip process env",
+      code: "deepseek_api_key_in_process",
+    };
+  }
+  return {
+    level: "error",
+    message: "DEEPSEEK_API_KEY not found in adapter env, profile .env, or process env",
+    hint: "Add DEEPSEEK_API_KEY to <HERMES_HOME>/.env or to the agent's env secrets.",
+    code: "deepseek_api_key_missing",
+  };
+}
+
+export async function testEnvironment(ctx) {
+  const config = ctx.config ?? {};
+  const command = asString(config.hermesCommand) || HERMES_CLI;
+  const profileHome = asString(config.hermesProfileHome) || DEFAULT_PROFILE_HOME;
+  const checks = [];
+
+  const cliCheck = await checkCliInstalled(command);
+  if (cliCheck) {
+    checks.push(cliCheck);
+    if (cliCheck.level === "error") {
+      return {
+        adapterType: ADAPTER_TYPE,
+        status: "fail",
+        checks,
+        testedAt: new Date().toISOString(),
+      };
+    }
+  }
+
+  const profileCheck = await checkProfile(profileHome);
+  checks.push(profileCheck);
+  if (profileCheck.level === "error") {
+    return {
+      adapterType: ADAPTER_TYPE,
+      status: "fail",
+      checks,
+      testedAt: new Date().toISOString(),
+    };
+  }
+
+  const apiKeyCheck = await checkApiKey(profileHome, config.env);
+  checks.push(apiKeyCheck);
+
+  const model = asString(config.model);
+  checks.push({
+    level: "info",
+    message: model ? `Model: ${model}` : "Using profile default model",
+    code: "deepseek_model",
+  });
+
+  const hasErrors = checks.some((c) => c.level === "error");
+  const hasWarnings = checks.some((c) => c.level === "warn");
+  return {
+    adapterType: ADAPTER_TYPE,
+    status: hasErrors ? "fail" : hasWarnings ? "warn" : "pass",
+    checks,
+    testedAt: new Date().toISOString(),
+  };
+}

adapters/deepseek-paperclip-adapter/dist/shared/constants.js

@@ -0,0 +1,36 @@
+/**
+ * Shared constants for the DeepSeek (via Hermes) Paperclip adapter.
+ */
+
+export const ADAPTER_TYPE = "deepseek_local";
+export const ADAPTER_LABEL = "DeepSeek (via Hermes)";
+
+/** Default Hermes CLI binary name. */
+export const HERMES_CLI = "hermes";
+
+/** Default profile directory used as HERMES_HOME if the agent does not override it. */
+export const DEFAULT_PROFILE_HOME = "/home/chaim/.hermes/profiles/deepseek";
+
+/** Default model — V4-Pro is the strongest DeepSeek model currently exposed. */
+export const DEFAULT_MODEL = "deepseek-v4-pro";
+
+/** DeepSeek profiles in this stack use Hermes' "custom" provider (user-defined in profile config.yaml). */
+export const DEFAULT_PROVIDER = "custom";
+
+/** Default timeout (seconds) for one CLI invocation. */
+export const DEFAULT_TIMEOUT_SEC = 1800;
+
+/** Grace period (seconds) after SIGTERM before SIGKILL. */
+export const DEFAULT_GRACE_SEC = 30;
+
+/** Models that DeepSeek's API currently exposes (verified via /v1/models). */
+export const DEEPSEEK_MODELS = [
+  { id: "deepseek-v4-pro",   label: "DeepSeek V4 Pro" },
+  { id: "deepseek-v4-flash", label: "DeepSeek V4 Flash" },
+];
+
+/** Regex for extracting session_id from quiet-mode Hermes output. */
+export const SESSION_ID_REGEX = /^session_id:\s*(\S+)/m;
+export const SESSION_ID_REGEX_LEGACY = /session[_ ](?:id|saved)[:\s]+([a-zA-Z0-9_-]+)/i;
+export const TOKEN_USAGE_REGEX = /tokens?[:\s]+(\d+)\s*(?:input|in)\b.*?(\d+)\s*(?:output|out)\b/i;
+export const COST_REGEX = /(?:cost|spent)[:\s]*\$?([\d.]+)/i;

adapters/deepseek-paperclip-adapter/package-lock.json

@@ -0,0 +1,25 @@
+{
+  "name": "deepseek-paperclip-adapter",
+  "version": "0.1.0",
+  "lockfileVersion": 3,
+  "requires": true,
+  "packages": {
+    "": {
+      "name": "deepseek-paperclip-adapter",
+      "version": "0.1.0",
+      "license": "MIT",
+      "dependencies": {
+        "@paperclipai/adapter-utils": "^2026.325.0"
+      },
+      "engines": {
+        "node": ">=20.0.0"
+      }
+    },
+    "node_modules/@paperclipai/adapter-utils": {
+      "version": "2026.428.0",
+      "resolved": "https://registry.npmjs.org/@paperclipai/adapter-utils/-/adapter-utils-2026.428.0.tgz",
+      "integrity": "sha512-kGHpE7rhePPCbnG3OwXbNuHZZuI+XyuFgNSiDnrEeiSbkI2c5XHM2WnWDCZ/NGHULfJW3lWhSxGMFoYqiy38vQ==",
+      "license": "MIT"
+    }
+  }
+}

adapters/deepseek-paperclip-adapter/package.json

@@ -0,0 +1,21 @@
+{
+  "name": "deepseek-paperclip-adapter",
+  "version": "0.1.0",
+  "description": "Paperclip adapter for DeepSeek (V4-Pro / V4-Flash) — runs Hermes Agent locally pinned to a DeepSeek profile",
+  "type": "module",
+  "license": "MIT",
+  "private": true,
+  "main": "./dist/index.js",
+  "exports": {
+    ".": "./dist/index.js"
+  },
+  "files": [
+    "dist"
+  ],
+  "dependencies": {
+    "@paperclipai/adapter-utils": "^2026.325.0"
+  },
+  "engines": {
+    "node": ">=20.0.0"
+  }
+}

docs/daphna-voice-fingerprint.md

@@ -400,6 +400,54 @@
 - **~30 תקדמים חיצוניים** ש**דפנה מצטטת באופן עקבי** (ראה precedent-network.md)
 - **~15 תקדמים אישיים** שלה עצמה — מהווים את הקאנון האישי שלה
 
+---
+
+
+## 6.11 לקחים מערר 1200-25 (קרית ענבים, מאי 2026)
+
+השוואה בין טיוטת הכותב לעריכת דפנה חשפה 7 דפוסי סגנון שלא היו מתועדים:
+
+### א. סדר בלוקים — תכניות לפני טענות (1xxx)
+בתיקי רישוי, דפנה מעדיפה שבלוק ט (תכניות חלות) יופיע **לפני** בלוק ז (טענות). הרציונל: הקורא צריך להכיר את המסגרת הנורמטיבית לפני שהוא קורא את טענות הצדדים.
+
+**סדר נכון ל-1xxx:** ה → ו → **ט** → ו.ב (רקע מורחב) → ז → ח → י → יא → יב
+
+### ב. תבנית "להלן מתוך" — חובה
+כל התייחסות למסמך מקור מלווה ב-"להלן מתוך [שם המסמך]:" כ-placeholder לציטוט/צילום. **12 מופעים** בעריכה, **0** בטיוטה. זהו דפוס סגנוני מרכזי שחייב להיות אוטומטי.
+
+דוגמאות:
+- "להלן מתוך הוראות התכנית:"
+- "להלן מתוך פרוטוקול הדיון בוועדה המקומית:"
+- "להלן מתוך הבקשה להיתר:"
+- "להלן מתוך מטרת התכנית:"
+- "להלן מתוך תשריט מצב מוצע:"
+
+### ג. רקע עובדתי מורחב — ציר זמן מלא
+בלוק ו חייב לספר את "הסיפור" של התיק: הגשת בקשה → פרסום → מספר התנגדויות → ישיבות ועדה מקומית (תאריך + תוצאה לכל אחת) → החלטה סופית → הגשת ערר. הטיוטה נתנה שורה אחת (90 מילים); דפנה הרחיבה ל-3 ישיבות מפורטות (~420 מילים).
+
+### ד. ניתוח "גשר תכנוני"
+כשמבקש שימוש חורג גם מקדם תכנית — דפנה מנתחת: האם השימוש המבוקש **תואם** את התכנון העתידי (→ גשר לגיטימי, כמו בכוכבה תורן)? או **סותר** (→ סטייה כפולה)? מסגרת ניתוח שלמה (249 מילים) שלא הייתה בטיוטה.
+
+### ה. עיגון כמותי
+דפנה מוסיפה נתונים מספריים ספציפיים: "4,404.98 מ"ר לכלל היישוב vs 1,425 מ"ר מבוקש — 32%". המספרים מעגנים את ההחלטה במציאות ומקשים על ערעור.
+
+### ו. כותרות שטוחות (Heading 2 בלבד)
+דפנה השתמשה ב-Heading 2 לכל הסעיפים, כולל תת-נושאים בדיון. **אין Heading 3**. כל סעיף עומד בפני עצמו.
+
+### ז. הבחנת תקדימים inline
+במקום סעיף נפרד "הבחנה מתקדימי העוררת" — ההבחנות מנוסחות inline: "באשר ל-[שם פסק דין]" → מה ההבדל → סיכום. דוגמה: "באשר לבג"ץ 6525/15 עמק שווה... אולם ההבדל מהותי".
+
+### ביטויי מעבר חדשים (מעריכה 1200-25)
+| ביטוי | הקשר |
+|-------|-------|
+| "עינינו הרואות" | ממצא מתוך מסמך |
+| "הנה כי כן" | לפיכך (פורמלי) |
+| "נשוב כאן ונבחין" | חזרה להבחנת תקדים |
+| "נוסיף ונבהיר" | הוספת הבהרה |
+| "מסקנת הדברים" | סיכום סעיף |
+| "משכבר קבענו" | הפניה לקביעה קודמת |
+
+
 ---
 
 ## 7. מה עדיין לא ראינו

docs/legal-decision-lessons.md

@@ -385,3 +385,64 @@ The draft's biggest structural error was adding the "נבאר" doctrinal paragra
 - [ ] Update voice-fingerprint: add new transition phrases
 - [ ] Update architecture-by-outcome: add "clean acceptance" archetype
 - [ ] Fix agent opening punctuation: "ונפרט;" not "נפרט."
+
+---
+
+## Lessons from ערר 1200-25 (קרית ענבים — שימוש חורג, דחייה)
+
+### Source
+- Our draft: `data/cases/1200-25/exports/טיוטה-v1.docx` (3,181 words)
+- Daphna's edit: `data/cases/1200-25/exports/עריכה-v1.docx` (4,313 words, +35%)
+- Date: May 2026
+
+### What the Edit Changed
+
+#### 1. Block Order — Plans Before Claims
+- **Draft:** ה→ו→ז→ח→ט→י→יא→יב (plans after procedures)
+- **Edit:** ה→ו→**ט**→ו.ב→ז→ח→י→יא→יב (plans BEFORE claims)
+- **Lesson:** In licensing cases (1xxx), the reader must understand the normative framework (plans) before reading the parties' arguments about those plans. Block ט should precede Block ז. The new order: opening → brief background → **applicable plans** → expanded background (application + committee proceedings) → claims → procedures → discussion.
+
+#### 2. "להלן מתוך" Document Insertion Pattern
+- **Draft:** 0 occurrences
+- **Edit:** 12 occurrences of "להלן מתוך [document name]:"
+- **Lesson:** Every reference to a source document must be accompanied by "להלן מתוך [שם המסמך]:" as a placeholder for a direct quote/image. This is a MANDATORY pattern, not optional. Examples: "להלן מתוך הוראות התכנית:", "להלן מתוך פרוטוקול הדיון:", "להלן מתוך הבקשה להיתר:"
+
+#### 3. Expanded Factual Background (Block ו)
+- **Draft:** ~90 words (3%), one paragraph
+- **Edit:** ~420 words (10%), covering: (a) the application details, (b) 3 committee meetings with dates and outcomes, (c) the final decision
+- **Lesson:** Block ו must tell the full "story" of the case: when the application was filed → when it was published → how many objections → when committee meetings were held → what was decided at each meeting → when the appeal was filed. Each meeting should have date + outcome.
+
+#### 4. Bridge Planning Analysis ("גשר תכנוני")
+- **Draft:** Not present
+- **Edit:** 249 words — new analytical framework
+- **Lesson:** When an applicant for deviation/variance is also promoting a plan for the same land, the decision must analyze: (a) is the pending plan harmonious with the requested use? If yes → the deviation can serve as a "bridge" until the plan is approved (cite כוכבה תורן). If no → the contradiction STRENGTHENS the rejection. The writer must check `search_case_documents` for pending plans and compare them with the requested use.
+
+#### 5. Competing Plans Analysis
+- **Draft:** Not present (1,033 words added)
+- **Edit:** Detailed comparison of the site-specific plan (151-1382787) vs the comprehensive plan (151-1337534)
+- **Lesson:** When there's a site-specific plan AND a comprehensive plan, the decision must: (a) describe each plan's scope, (b) compare the permitted uses, (c) show quantitative contradictions (e.g., "the comprehensive plan allocates 4,404 m² for ALL commerce in the settlement, while the request alone is for 1,425 m² — 32%"), (d) conclude whether there's harmony or contradiction. This is often the STRONGEST argument in the decision.
+
+#### 6. Heading Level — Flat Structure
+- **Draft:** Mixed Heading 2 + Heading 3 (nested subsections)
+- **Edit:** All Heading 2 (flat structure)
+- **Lesson:** Each section stands independently. No nesting. In the discussion, each analytical step is a separate Heading 2 section.
+
+#### 7. Inline Precedent Distinguishing
+- **Draft:** Separate section "הבחנה מתקדימי העוררת" (Heading 3)
+- **Edit:** Each precedent distinguished inline with "באשר ל-[case name]" → what's different → conclusion
+- **Lesson:** Don't create a separate "distinguishing" section. Address each precedent where it naturally comes up in the discussion, using "באשר ל..." as the opener.
+
+### New Transition Phrases Identified
+- **"עינינו הרואות"** — introducing a document-based finding ("our eyes see that...")
+- **"הנה כי כן"** — therefore/accordingly (more formal than "לפיכך")
+- **"נשוב כאן ונבחין"** — returning to distinguish a case
+- **"נוסיף ונבהיר"** — adding clarification
+- **"מסקנת הדברים"** — concluding a subsection
+- **"משכבר קבענו"** — since we already established
+
+### Applied To
+- [x] Update legal-decision-lessons.md with lessons 1-7
+- [x] Update daphna-voice-fingerprint.md with structural and style findings
+- [ ] Update block-schema.md: block order for 1xxx cases (ט before ז)
+- [ ] Update daphna-architecture-by-outcome.md: add "bridge planning" analysis for rejections
+- [ ] Update writer system prompt: mandatory "להלן מתוך" pattern

mcp-server/src/legal_mcp/server.py

@@ -116,6 +116,13 @@ async def case_delete(case_number: str, remove_files: bool = False) -> str:
     return await cases.case_delete(case_number, remove_files)
 
 
+@mcp.tool()
+async def case_get_final_text(case_number: str, max_chars: int = 0) -> str:
+    """קליטת טקסט ההחלטה הסופית (`סופי-{case}.docx` בתיקיית exports).
+    max_chars: 0=הכל, אחרת חיתוך לאורך הנתון. שימושי ל-Hermes Knowledge Curator."""
+    return await cases.case_get_final_text(case_number, max_chars)
+
+
 # Precedent attachments (user-supplied legal support for the compose phase)
 @mcp.tool()
 async def precedent_attach(
@@ -209,6 +216,22 @@ async def precedent_library_delete(case_law_id: str) -> str:
     return await plib.precedent_library_delete(case_law_id)
 
 
+@mcp.tool()
+async def precedent_link_cases(
+    case_law_id_a: str,
+    case_law_id_b: str,
+    relation_type: str = "same_case_chain",
+) -> str:
+    """קישור שתי פסיקות כקשורות (דו-כיווני, idempotent). relation_type: same_case_chain | overruled_by | distinguished."""
+    return await plib.precedent_link_cases(case_law_id_a, case_law_id_b, relation_type)
+
+
+@mcp.tool()
+async def precedent_unlink_cases(case_law_id_a: str, case_law_id_b: str) -> str:
+    """הסרת קישור בין שתי פסיקות (דו-כיווני)."""
+    return await plib.precedent_unlink_cases(case_law_id_a, case_law_id_b)
+
+
 @mcp.tool()
 async def precedent_extract_halachot(case_law_id: str) -> str:
     """הרצה מחדש של חילוץ הלכות לפסיקה קיימת. ההלכות הקיימות נמחקות, החדשות חוזרות לסטטוס pending_review."""
@@ -390,6 +413,35 @@ async def find_similar_cases(
     )
 
 
+@mcp.tool()
+async def search_internal_decisions(
+    query: str,
+    practice_area: str = "",
+    appeal_subtype: str = "",
+    district: str = "",
+    chair_name: str = "",
+    limit: int = 10,
+    include_halachot: bool = True,
+) -> str:
+    """חיפוש בהחלטות ועדות ערר לתכנון ובנייה (כל המחוזות).
+
+    מחזיר החלטות מהקורפוס הפנימי של ועדות הערר — נפרד מפסיקת בתי המשפט.
+    השתמש בו במקביל ל-search_precedent_library להצגת שתי שכבות נפרדות.
+
+    Args:
+        query: שאילתת חיפוש בעברית
+        practice_area: rishuy_uvniya / betterment_levy / compensation_197
+        appeal_subtype: סינון לפי תת-סוג ערר
+        district: מחוז — ירושלים / מרכז / תל אביב / צפון / דרום / ארצי. ריק = כל המחוזות
+        chair_name: שם יו"ר הוועדה לסינון. ריק = כל היו"רים
+        limit: מספר תוצאות מקסימלי
+        include_halachot: האם לכלול הלכות שחולצו
+    """
+    return await search.search_internal_decisions(
+        query, practice_area, appeal_subtype, district, chair_name, limit, include_halachot,
+    )
+
+
 # Drafting
 @mcp.tool()
 async def get_style_guide() -> str:
@@ -573,6 +625,43 @@ async def ingest_final_version(
     return await workflow.ingest_final_version(case_number, file_path, final_text)
 
 
+@mcp.tool()
+async def internal_decision_migrate(
+    source: str = "both",
+    dry_run: bool = True,
+) -> str:
+    """העברת החלטות ועדת ערר קיימות לקורפוס הפנימי (פעולת admin).
+
+    source: 'style_corpus' | 'external_corpus' | 'both'
+    dry_run: אם true — מציג מה יקרה ללא כתיבה
+    """
+    import json as _json
+    from legal_mcp.services import internal_decisions as int_svc
+    if source not in {"style_corpus", "external_corpus", "both"}:
+        return "source חייב להיות style_corpus / external_corpus / both"
+    results: dict = {}
+    if source in {"style_corpus", "both"}:
+        results["style_corpus"] = await int_svc.migrate_from_style_corpus(dry_run=dry_run)
+    if source in {"external_corpus", "both"}:
+        results["external_corpus"] = await int_svc.migrate_from_external_corpus(dry_run=dry_run)
+    return _json.dumps(results, ensure_ascii=False, indent=2)
+
+
+@mcp.tool()
+async def internal_decision_enrich(
+    dry_run: bool = True,
+) -> str:
+    """העשרת החלטות שהומגרו (חד-פעמי): תיקון מספר ערר + שם + תאריך + תור להלכות.
+
+    dry_run=True — מציג כמה רשומות יטופלו ללא כתיבה.
+    dry_run=False — מריץ בפועל: metadata extraction (תיקון case_number/case_name/date) ואחר כך תור חילוץ הלכות.
+    """
+    import json as _json
+    from legal_mcp.services import internal_decisions as int_svc
+    result = await int_svc.enrich_migrated_entries(dry_run=dry_run)
+    return _json.dumps(result, ensure_ascii=False, indent=2)
+
+
 @mcp.tool()
 async def record_chair_feedback(
     case_number: str,

mcp-server/src/legal_mcp/services/block_writer.py

@@ -360,13 +360,9 @@ async def write_block(
         post_hearing_context=post_hearing_context,
     )
 
-    # Restructure: sources first, then instructions
+    # source_context is already embedded inside formatted_prompt via {source_context} in the
-    prompt = (
+    # template. Do NOT prepend it again — doing so doubles the prompt size (was 465K chars).
-        f"## חומרי מקור (מסמכים מלאים — צטט מהם מילה במילה כשאפשר):\n\n"
+    prompt = formatted_prompt
-        f"{source_context}\n\n"
-        f"---\n\n"
-        f"{formatted_prompt}"
-    )
 
     if instructions:
         prompt += f"\n\n## הנחיות נוספות:\n{instructions}"
@@ -377,6 +373,19 @@ async def write_block(
         if not dir_doc.get("approved"):
             raise ValueError("לא ניתן לכתוב בלוק דיון ללא כיוון מאושר. הפעל brainstorm → approve_direction קודם.")
 
+    # Guard against context overflow before calling claude -p.
+    # Sonnet: 200K context → ~800K chars max; Opus: 200K context → same.
+    # In practice the CLI has crashed on prompts above ~400K chars, so use
+    # that as a conservative ceiling (well below the token limit).
+    _MAX_PROMPT_CHARS = 400_000
+    if len(prompt) > _MAX_PROMPT_CHARS:
+        raise RuntimeError(
+            f"Prompt too large for {block_id}: {len(prompt):,} chars "
+            f"(limit {_MAX_PROMPT_CHARS:,}). "
+            f"source_context: {len(source_context):,} chars. "
+            f"Reduce documents or call extract_appraiser_facts first."
+        )
+
     # Call Claude via Claude Code session (no API)
     model_key = block_cfg["model"]
     timeout = claude_session.LONG_TIMEOUT if model_key == "opus" else claude_session.DEFAULT_TIMEOUT
@@ -414,16 +423,35 @@ def _build_case_context(case: dict, decision: dict | None) -> str:
 - תוצאה: {outcome_heb}"""
 
 
+# Which doc_types are relevant per block.
+# None  → skip source docs entirely (block uses other context, e.g. claims_context)
+# []    → include all doc types (default for unspecified blocks)
+# [..] → include only the listed doc_type values
+_BLOCK_DOC_TYPES: dict[str, list[str] | None] = {
+    "block-he":    None,                      # only case_context needed; no full docs
+    "block-vav":   ["appeal", "protocol"],    # כתב ערר + פרוטוקול ועדה
+    "block-zayin": None,                      # claims_context is sufficient
+    "block-chet":  ["protocol"],              # פרוטוקול + השלמות טיעון
+    "block-tet":   ["appraisal"],             # שומות בלבד
+    # block-yod, block-yod-alef, block-he etc. default → all docs
+}
+
+
 async def _build_source_context(case_id: UUID, block_id: str) -> str:
-    """Get full document texts for the block.
+    """Get document texts for the block, filtered by relevance.
 
     Per Anthropic best practices: send full source documents, not truncated excerpts.
-    Place documents at the TOP of the prompt (before instructions) for 30% better recall.
+    Per-block filtering prevents context overflow on large cases (9+ docs).
-    For grounding: instruct Claude to cite word-for-word from these documents.
     """
+    allowed = _BLOCK_DOC_TYPES.get(block_id, [])  # [] sentinel = not in map → all docs
+    if allowed is None:
+        return ""  # this block doesn't need raw source docs
+
     docs = await db.list_documents(case_id)
     context_parts = []
     for doc in docs:
+        if allowed and doc["doc_type"] not in allowed:
+            continue
         text = await db.get_document_text(UUID(doc["id"]))
         if text:
             context_parts.append(f"--- מסמך: {doc['title']} ({doc['doc_type']}) ---\n{text}")

mcp-server/src/legal_mcp/services/claude_session.py

@@ -72,6 +72,9 @@ async def query(
     """
     full_prompt = f"{system}\n\n{prompt}" if system else prompt
 
+    if len(full_prompt) > 150_000:
+        logger.warning("Large prompt: %d chars — may hit context limits", len(full_prompt))
+
     cmd = [
         "claude", "-p",
         "--output-format", "json",
@@ -110,7 +113,8 @@ async def query(
 
     if proc.returncode != 0:
         stderr = stderr_b.decode("utf-8", errors="replace").strip()[:500] or "unknown error"
-        raise RuntimeError(f"Claude CLI failed (exit {proc.returncode}): {stderr}")
+        size_info = f"; prompt_len={len(full_prompt):,} chars" if len(full_prompt) > 100_000 else ""
+        raise RuntimeError(f"Claude CLI failed (exit {proc.returncode}): {stderr}{size_info}")
 
     stdout = stdout_b.decode("utf-8", errors="replace").strip()
     if not stdout:

mcp-server/src/legal_mcp/services/db.py

@@ -691,6 +691,30 @@ CREATE INDEX IF NOT EXISTS idx_prec_img_emb_case_law
 """
 
 
+SCHEMA_V10_SQL = """
+ALTER TABLE case_law ADD COLUMN IF NOT EXISTS chair_name TEXT DEFAULT '';
+ALTER TABLE case_law ADD COLUMN IF NOT EXISTS district TEXT DEFAULT '';
+ALTER TABLE cases ADD COLUMN IF NOT EXISTS chair_name TEXT DEFAULT '';
+CREATE INDEX IF NOT EXISTS idx_case_law_source_kind ON case_law(source_kind);
+CREATE INDEX IF NOT EXISTS idx_case_law_chair ON case_law(chair_name) WHERE chair_name <> '';
+CREATE INDEX IF NOT EXISTS idx_case_law_district ON case_law(district) WHERE district <> '';
+"""
+
+SCHEMA_V11_SQL = """
+CREATE TABLE IF NOT EXISTS case_law_relations (
+    id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
+    case_law_id UUID NOT NULL REFERENCES case_law(id) ON DELETE CASCADE,
+    related_id   UUID NOT NULL REFERENCES case_law(id) ON DELETE CASCADE,
+    relation_type TEXT NOT NULL DEFAULT 'same_case_chain',
+    created_at TIMESTAMPTZ DEFAULT now(),
+    UNIQUE(case_law_id, related_id),
+    CHECK (case_law_id <> related_id)
+);
+CREATE INDEX IF NOT EXISTS idx_clr_a ON case_law_relations(case_law_id);
+CREATE INDEX IF NOT EXISTS idx_clr_b ON case_law_relations(related_id);
+"""
+
+
 async def _run_schema_migrations(pool: asyncpg.Pool) -> None:
     async with pool.acquire() as conn:
         await conn.execute(SCHEMA_SQL)
@@ -703,7 +727,9 @@ async def _run_schema_migrations(pool: asyncpg.Pool) -> None:
         await conn.execute(SCHEMA_V7_SQL)
         await conn.execute(SCHEMA_V8_SQL)
         await conn.execute(SCHEMA_V9_SQL)
-    logger.info("Database schema initialized (v1-v9)")
+        await conn.execute(SCHEMA_V10_SQL)
+        await conn.execute(SCHEMA_V11_SQL)
+    logger.info("Database schema initialized (v1-v11)")
 
 
 async def init_schema() -> None:
@@ -1724,6 +1750,59 @@ async def get_case_law(case_law_id: UUID) -> dict | None:
     return _row_to_case_law(row) if row else None
 
 
+async def add_case_law_relation(
+    a_id: UUID, b_id: UUID, relation_type: str = "same_case_chain"
+) -> None:
+    """Link two case_law records bidirectionally. Idempotent (ON CONFLICT DO NOTHING)."""
+    pool = await get_pool()
+    async with pool.acquire() as conn:
+        await conn.executemany(
+            """
+            INSERT INTO case_law_relations(case_law_id, related_id, relation_type)
+            VALUES($1, $2, $3)
+            ON CONFLICT (case_law_id, related_id) DO NOTHING
+            """,
+            [(a_id, b_id, relation_type), (b_id, a_id, relation_type)],
+        )
+
+
+async def remove_case_law_relation(a_id: UUID, b_id: UUID) -> None:
+    """Remove a bidirectional link between two case_law records."""
+    pool = await get_pool()
+    await pool.execute(
+        """
+        DELETE FROM case_law_relations
+        WHERE (case_law_id = $1 AND related_id = $2)
+           OR (case_law_id = $2 AND related_id = $1)
+        """,
+        a_id,
+        b_id,
+    )
+
+
+async def get_case_law_relations(case_law_id: UUID) -> list[dict]:
+    """Return all case_law records linked to case_law_id, ordered by date asc."""
+    pool = await get_pool()
+    rows = await pool.fetch(
+        """
+        SELECT cl.*, r.relation_type
+        FROM case_law_relations r
+        JOIN case_law cl ON cl.id = r.related_id
+        WHERE r.case_law_id = $1
+        ORDER BY cl.date ASC NULLS LAST
+        """,
+        case_law_id,
+    )
+    results = []
+    for row in rows:
+        d = dict(row)
+        relation_type = d.pop("relation_type")
+        normalized = _row_to_case_law(d)
+        normalized["relation_type"] = relation_type
+        results.append(normalized)
+    return results
+
+
 async def get_case_law_by_citation(case_number: str) -> dict | None:
     pool = await get_pool()
     row = await pool.fetchrow(
@@ -1817,6 +1896,85 @@ async def create_external_case_law(
     return _row_to_case_law(row)
 
 
+async def create_internal_committee_decision(
+    case_number: str,
+    case_name: str,
+    full_text: str,
+    court: str = "",
+    decision_date: date | None = None,
+    chair_name: str = "",
+    district: str = "",
+    practice_area: str = "",
+    appeal_subtype: str = "",
+    subject_tags: list[str] | None = None,
+    summary: str = "",
+    is_binding: bool = True,
+    document_id: UUID | None = None,
+) -> dict:
+    """Upsert an appeals-committee decision as source_kind='internal_committee'.
+
+    If a row with this case_number already exists as cited_only, promotes it.
+    Idempotent: calling again updates metadata in-place.
+    """
+    pool = await get_pool()
+    tags_json = json.dumps(subject_tags or [], ensure_ascii=False)
+    async with pool.acquire() as conn:
+        existing = await conn.fetchrow(
+            "SELECT id FROM case_law WHERE case_number = $1",
+            case_number,
+        )
+        if existing:
+            row = await conn.fetchrow(
+                """
+                UPDATE case_law SET
+                    case_name = $2,
+                    court = COALESCE(NULLIF($3, ''), court),
+                    date = COALESCE($4, date),
+                    chair_name = COALESCE(NULLIF($5, ''), chair_name),
+                    district = COALESCE(NULLIF($6, ''), district),
+                    practice_area = $7,
+                    appeal_subtype = $8,
+                    subject_tags = $9,
+                    summary = COALESCE(NULLIF($10, ''), summary),
+                    full_text = $11,
+                    source_type = 'appeals_committee',
+                    source_kind = 'internal_committee',
+                    is_binding = $12,
+                    document_id = COALESCE($13, document_id),
+                    extraction_status = 'processing',
+                    halacha_extraction_status = 'pending'
+                WHERE id = $1
+                RETURNING *
+                """,
+                existing["id"], case_name, court, decision_date,
+                chair_name, district, practice_area, appeal_subtype,
+                tags_json, summary, full_text, is_binding, document_id,
+            )
+        else:
+            row = await conn.fetchrow(
+                """
+                INSERT INTO case_law (
+                    case_number, case_name, court, date, chair_name, district,
+                    subject_tags, summary, full_text,
+                    source_kind, source_type, document_id,
+                    extraction_status, halacha_extraction_status,
+                    practice_area, appeal_subtype, is_binding
+                ) VALUES (
+                    $1, $2, $3, $4, $5, $6,
+                    $7, $8, $9,
+                    'internal_committee', 'appeals_committee', $10,
+                    'processing', 'pending',
+                    $11, $12, $13
+                )
+                RETURNING *
+                """,
+                case_number, case_name, court, decision_date, chair_name, district,
+                tags_json, summary, full_text,
+                document_id, practice_area, appeal_subtype, is_binding,
+            )
+    return _row_to_case_law(row)
+
+
 async def update_case_law(case_law_id: UUID, **fields) -> dict | None:
     """Patch metadata fields on a case_law row.
 
@@ -1825,7 +1983,7 @@ async def update_case_law(case_law_id: UUID, **fields) -> dict | None:
     precedent_level, is_binding.
     """
     allowed = {
-        "case_name", "court", "date", "practice_area", "appeal_subtype",
+        "case_number", "case_name", "court", "date", "practice_area", "appeal_subtype",
         "subject_tags", "summary", "headnote", "key_quote", "source_url",
         "source_type", "precedent_level", "is_binding",
     }
@@ -1892,10 +2050,21 @@ async def list_external_case_law(
     search: str = "",
     limit: int = 100,
     offset: int = 0,
+    source_kind: str = "external_upload",
 ) -> list[dict]:
-    """List chair-uploaded precedents, with simple filters."""
+    """List chair-uploaded precedents, with simple filters.
+
+    source_kind="all_committees" expands to: source_kind='internal_committee'
+    OR (source_kind='external_upload' AND source_type='appeals_committee').
+    """
     pool = await get_pool()
-    conditions = ["source_kind = 'external_upload'"]
+    if source_kind == "all_committees":
+        conditions = [
+            "(source_kind = 'internal_committee' OR "
+            "(source_kind = 'external_upload' AND source_type = 'appeals_committee'))"
+        ]
+    else:
+        conditions = [f"source_kind = '{source_kind}'"]
     params: list = []
     idx = 1
     if practice_area:
@@ -1927,6 +2096,7 @@ async def list_external_case_law(
         SELECT id, case_number, case_name, court, date, practice_area,
                appeal_subtype, source_type, precedent_level, is_binding,
                summary, headnote, subject_tags, source_kind,
+               chair_name, district,
                extraction_status, halacha_extraction_status,
                metadata_extraction_requested_at,
                halacha_extraction_requested_at,
@@ -2194,8 +2364,14 @@ async def search_precedent_library_semantic(
     subject_tag: str = "",
     limit: int = 10,
     include_halachot: bool = True,
+    source_kind: str = "external_upload",
+    district: str = "",
+    chair_name: str = "",
 ) -> list[dict]:
-    """Semantic search over chair-uploaded precedents.
+    """Semantic search over precedents filtered by source_kind.
+
+    source_kind='external_upload'  → court rulings (default)
+    source_kind='internal_committee' → appeals-committee decisions
 
     Returns merged halachot + chunks. Halachot are pre-distilled rules, so
     they get a small score boost. Only ``approved`` / ``published`` halachot
@@ -2204,7 +2380,7 @@ async def search_precedent_library_semantic(
     """
     pool = await get_pool()
     halacha_filters = ["h.review_status IN ('approved', 'published')"]
-    chunk_filters = ["cl.source_kind = 'external_upload'"]
+    chunk_filters = [f"cl.source_kind = '{source_kind}'"]
     h_params: list = [query_embedding, limit]
     c_params: list = [query_embedding, limit]
     h_idx = 3
@@ -2249,13 +2425,27 @@ async def search_precedent_library_semantic(
         halacha_filters.append(f"${h_idx} = ANY(h.subject_tags)")
         h_params.append(subject_tag)
         h_idx += 1
+    if district:
+        halacha_filters.append(f"cl.district = ${h_idx}")
+        h_params.append(district)
+        h_idx += 1
+        chunk_filters.append(f"cl.district = ${c_idx}")
+        c_params.append(district)
+        c_idx += 1
+    if chair_name:
+        halacha_filters.append(f"cl.chair_name = ${h_idx}")
+        h_params.append(chair_name)
+        h_idx += 1
+        chunk_filters.append(f"cl.chair_name = ${c_idx}")
+        c_params.append(chair_name)
+        c_idx += 1
 
     halacha_sql = f"""
         SELECT h.id AS halacha_id, h.case_law_id, h.rule_statement,
                h.reasoning_summary, h.supporting_quote, h.page_reference,
                h.practice_areas, h.subject_tags, h.confidence, h.rule_type,
                cl.case_number, cl.case_name, cl.court, cl.date AS decision_date,
-               cl.precedent_level,
+               cl.precedent_level, cl.chair_name, cl.district,
                1 - (h.embedding <=> $1) AS score
         FROM halachot h
         JOIN case_law cl ON cl.id = h.case_law_id
@@ -2269,7 +2459,7 @@ async def search_precedent_library_semantic(
         SELECT pc.id AS chunk_id, pc.case_law_id, pc.content,
                pc.section_type, pc.page_number,
                cl.case_number, cl.case_name, cl.court, cl.date AS decision_date,
-               cl.precedent_level, cl.practice_area,
+               cl.precedent_level, cl.practice_area, cl.chair_name, cl.district,
                1 - (pc.embedding <=> $1) AS score
         FROM precedent_chunks pc
         JOIN case_law cl ON cl.id = pc.case_law_id
@@ -2308,19 +2498,17 @@ async def precedent_library_stats() -> dict:
     pool = await get_pool()
     async with pool.acquire() as conn:
         total = await conn.fetchval(
-            "SELECT COUNT(*) FROM case_law WHERE source_kind = 'external_upload'"
+            "SELECT COUNT(*) FROM case_law"
         )
         by_practice = await conn.fetch(
             """SELECT practice_area, COUNT(*) AS n
                FROM case_law
-               WHERE source_kind = 'external_upload'
                GROUP BY practice_area
                ORDER BY n DESC"""
         )
         by_level = await conn.fetch(
             """SELECT precedent_level, COUNT(*) AS n
                FROM case_law
-               WHERE source_kind = 'external_upload'
                GROUP BY precedent_level
                ORDER BY n DESC"""
         )
@@ -2354,22 +2542,31 @@ async def precedent_library_stats() -> dict:
 
 async def request_metadata_extraction(case_law_id: UUID) -> bool:
     """Stamp ``metadata_extraction_requested_at`` for the local MCP worker
-    to pick up. Returns False if the row is missing."""
+    to pick up. Returns False if the row is missing.
+
+    Originally restricted to ``source_kind='external_upload'`` (see git
+    blame). Opened to all source kinds 2026-05-06 — internal_committee
+    rows can also need re-extraction (e.g. corrupted subject_tags from
+    an early ingest pipeline). The extractor itself preserves user
+    values (``precedent_metadata_extractor.extract_and_apply`` only
+    fills empty fields), so this is safe.
+    """
     pool = await get_pool()
     result = await pool.execute(
         "UPDATE case_law SET metadata_extraction_requested_at = now() "
-        "WHERE id = $1 AND source_kind = 'external_upload'",
+        "WHERE id = $1",
         case_law_id,
     )
     return result == "UPDATE 1"
 
 
 async def request_halacha_extraction(case_law_id: UUID) -> bool:
-    """Same but for halacha extraction."""
+    """Same but for halacha extraction. See note on
+    :func:`request_metadata_extraction` re: opening to all source kinds."""
     pool = await get_pool()
     result = await pool.execute(
         "UPDATE case_law SET halacha_extraction_requested_at = now() "
-        "WHERE id = $1 AND source_kind = 'external_upload'",
+        "WHERE id = $1",
         case_law_id,
     )
     return result == "UPDATE 1"
@@ -2389,12 +2586,15 @@ async def list_pending_extraction_requests(
         else "halacha_extraction_requested_at"
     )
     pool = await get_pool()
+    # Drop the legacy ``source_kind = 'external_upload'`` filter — without it
+    # internal_committee rows could be stamped (we opened that gate in
+    # request_metadata_extraction / request_halacha_extraction) but stayed
+    # invisible to the worker forever.
     rows = await pool.fetch(
         f"""SELECT id, case_number, case_name, court, date,
                   practice_area, is_binding, {col} AS requested_at
             FROM case_law
             WHERE {col} IS NOT NULL
-              AND source_kind = 'external_upload'
             ORDER BY {col} ASC
             LIMIT $1""",
         limit,

mcp-server/src/legal_mcp/services/hybrid_search.py

@@ -88,8 +88,15 @@ async def search_precedent_library_hybrid(
     is_binding: bool | None = None,
     subject_tag: str = "",
     include_halachot: bool = True,
+    source_kind: str = "external_upload",
+    district: str = "",
+    chair_name: str = "",
 ) -> list[dict]:
-    """Hybrid wrapper for precedent-library search."""
+    """Hybrid wrapper for precedent-library search.
+
+    source_kind='external_upload'    → court rulings (default)
+    source_kind='internal_committee' → appeals-committee decisions
+    """
     fetch_k = max(limit, config.VOYAGE_RERANK_FETCH_K) if config.MULTIMODAL_ENABLED else limit
 
     async def _base(limit: int) -> list[dict]:
@@ -103,6 +110,9 @@ async def search_precedent_library_hybrid(
             subject_tag=subject_tag,
             limit=limit,
             include_halachot=include_halachot,
+            source_kind=source_kind,
+            district=district,
+            chair_name=chair_name,
         )
 
     text_results = await rerank.maybe_rerank(

mcp-server/src/legal_mcp/services/internal_decisions.py

@@ -0,0 +1,376 @@
+"""Orchestrator for the Internal Committee Decisions corpus.
+
+Ingest pipeline:
+    text/file → INSERT case_law (source_kind='internal_committee')
+        → chunk → embed → store precedent_chunks
+        → queue halacha extraction
+
+Migration helpers:
+    migrate_from_style_corpus()   — re-index style_corpus entries as searchable
+    migrate_from_external_corpus() — reclassify external appeals-committee rows
+
+All ועדות ערר (any district) belong here.
+Judicial decisions (Supreme Court, Administrative Court) stay in external_upload.
+"""
+
+from __future__ import annotations
+
+import logging
+import re
+import shutil
+from datetime import date
+from pathlib import Path
+from uuid import UUID, uuid4
+
+from legal_mcp import config
+from legal_mcp.services import chunker, db, embeddings, extractor
+
+logger = logging.getLogger(__name__)
+
+INTERNAL_DECISIONS_DIR = Path(config.DATA_DIR) / "internal-decisions"
+
+_VALID_DISTRICTS = {"", "ירושלים", "מרכז", "תל אביב", "צפון", "דרום", "ארצי"}
+
+_COURT_TO_DISTRICT = [
+    ("ירושלים", "ירושלים"),
+    ("תל אביב", "תל אביב"),
+    ('ת"א', "תל אביב"),
+    ("מרכז", "מרכז"),
+    ("חיפה", "צפון"),
+    ("צפון", "צפון"),
+    ("דרום", "דרום"),
+    ("ארצי", "ארצי"),
+    ("ארצית", "ארצי"),
+]
+
+
+def _coerce_date(value) -> date | None:
+    if value is None or value == "":
+        return None
+    if isinstance(value, date):
+        return value
+    if isinstance(value, str):
+        try:
+            return date.fromisoformat(value[:10])
+        except ValueError:
+            return None
+    return None
+
+
+def _safe_filename(name: str) -> str:
+    base = Path(name).name
+    return re.sub(r"[^\w.\-+א-ת ]", "_", base) or f"internal-{uuid4().hex[:8]}"
+
+
+def _district_from_court(court: str) -> str:
+    for keyword, district in _COURT_TO_DISTRICT:
+        if keyword in court:
+            return district
+    return ""
+
+
+async def ingest_internal_decision(
+    *,
+    case_number: str,
+    case_name: str = "",
+    court: str = "",
+    decision_date=None,
+    chair_name: str = "",
+    district: str = "",
+    practice_area: str = "",
+    appeal_subtype: str = "",
+    subject_tags: list[str] | None = None,
+    summary: str = "",
+    is_binding: bool = True,
+    file_path: str | Path | None = None,
+    text: str | None = None,
+    document_id: UUID | None = None,
+    queue_halachot: bool = True,
+) -> dict:
+    """Ingest an appeals-committee decision into the internal corpus.
+
+    Either file_path or text must be provided.
+    If district is empty, it is inferred from court.
+    Returns: {"status": "completed", "case_law_id": "...", "chunks": N}
+    """
+    if not file_path and not text:
+        raise ValueError("either file_path or text is required")
+    if not case_number.strip():
+        raise ValueError("case_number is required")
+
+    resolved_district = district.strip() or _district_from_court(court)
+
+    if file_path:
+        src = Path(file_path)
+        if not src.is_file():
+            raise FileNotFoundError(f"file not found: {src}")
+        dest_dir = INTERNAL_DECISIONS_DIR / (resolved_district or "other")
+        dest_dir.mkdir(parents=True, exist_ok=True)
+        staged = dest_dir / f"{uuid4().hex[:8]}_{_safe_filename(src.name)}"
+        shutil.copy2(src, staged)
+        raw_text, page_count, page_offsets = await extractor.extract_text(str(staged))
+        raw_text = extractor.strip_nevo_preamble(raw_text or "").strip()
+        if not raw_text:
+            raise ValueError("no extractable text in file")
+    else:
+        raw_text = (text or "").strip()
+        if not raw_text:
+            raise ValueError("text is empty")
+        page_count = 0
+        page_offsets = None
+
+    record = await db.create_internal_committee_decision(
+        case_number=case_number.strip(),
+        case_name=(case_name.strip() or case_number.strip()),
+        full_text=raw_text,
+        court=court.strip(),
+        decision_date=_coerce_date(decision_date),
+        chair_name=chair_name.strip(),
+        district=resolved_district,
+        practice_area=practice_area,
+        appeal_subtype=appeal_subtype.strip(),
+        subject_tags=list(subject_tags or []),
+        summary=summary.strip(),
+        is_binding=is_binding,
+        document_id=document_id,
+    )
+    case_law_id = UUID(str(record["id"]))
+
+    try:
+        chunks = chunker.chunk_document(raw_text, page_offsets=page_offsets)
+        if not chunks:
+            await db.set_case_law_extraction_status(case_law_id, "completed")
+            await db.set_case_law_halacha_status(case_law_id, "completed")
+            return {"status": "completed", "case_law_id": str(case_law_id), "chunks": 0}
+
+        chunk_texts = [c.content for c in chunks]
+        chunk_vectors = await embeddings.embed_texts(chunk_texts, input_type="document")
+        chunk_dicts = [
+            {
+                "chunk_index": c.chunk_index,
+                "content": c.content,
+                "section_type": c.section_type,
+                "page_number": c.page_number,
+                "embedding": v,
+            }
+            for c, v in zip(chunks, chunk_vectors)
+        ]
+        stored = await db.store_precedent_chunks(case_law_id, chunk_dicts)
+
+        await db.set_case_law_extraction_status(case_law_id, "completed")
+        await db.set_case_law_halacha_status(case_law_id, "pending")
+        if queue_halachot:
+            await db.request_halacha_extraction(case_law_id)
+
+        return {
+            "status": "completed",
+            "case_law_id": str(case_law_id),
+            "chunks": stored,
+            "halachot_pending": True,
+        }
+
+    except Exception:
+        logger.exception("ingest_internal_decision failed for %s", case_number)
+        await db.set_case_law_extraction_status(case_law_id, "failed")
+        raise
+
+
+async def migrate_from_style_corpus(dry_run: bool = False, queue_halachot: bool = True) -> dict:
+    """Re-index all style_corpus entries as searchable internal committee decisions.
+
+    Does NOT delete style_corpus rows — they remain for style analysis.
+    Skips entries that already exist in case_law as internal_committee.
+    """
+    pool = await db.get_pool()
+    async with pool.acquire() as conn:
+        rows = await conn.fetch(
+            """SELECT decision_number, decision_date, full_text,
+                      practice_area, appeal_subtype, subject_categories
+               FROM style_corpus
+               ORDER BY decision_date NULLS LAST"""
+        )
+
+    results = {"total": len(rows), "ingested": 0, "skipped": 0, "failed": 0, "dry_run": dry_run}
+
+    for row in rows:
+        case_number = (row["decision_number"] or "").strip()
+        if not case_number:
+            results["skipped"] += 1
+            continue
+
+        if not dry_run:
+            existing = await pool.fetchval(
+                "SELECT id FROM case_law WHERE case_number = $1 AND source_kind = 'internal_committee'",
+                case_number,
+            )
+            if existing:
+                results["skipped"] += 1
+                continue
+
+        if dry_run:
+            results["ingested"] += 1
+            continue
+
+        try:
+            subject_tags = list(row["subject_categories"] or [])
+            raw_pa = row["practice_area"] or ""
+            subtype = row["appeal_subtype"] or ""
+            # style_corpus stores 'appeals_committee' (source_type) instead of practice_area
+            _subtype_to_pa = {
+                "building_permit": "rishuy_uvniya",
+                "betterment_levy": "betterment_levy",
+                "compensation_197": "compensation_197",
+            }
+            practice_area = raw_pa if raw_pa in ("rishuy_uvniya", "betterment_levy", "compensation_197") \
+                else _subtype_to_pa.get(subtype, "")
+            await ingest_internal_decision(
+                case_number=case_number,
+                court="ועדת הערר לתכנון ובנייה — מחוז ירושלים",
+                decision_date=row["decision_date"],
+                chair_name="דפנה תמיר",
+                district="ירושלים",
+                practice_area=practice_area,
+                appeal_subtype=subtype,
+                subject_tags=subject_tags,
+                text=row["full_text"],
+                queue_halachot=queue_halachot,
+            )
+            results["ingested"] += 1
+            logger.info("Migrated style_corpus entry: %s", case_number)
+        except Exception as e:
+            logger.error("Failed to migrate %s: %s", case_number, e)
+            results["failed"] += 1
+
+    return results
+
+
+async def migrate_from_external_corpus(dry_run: bool = False) -> dict:
+    """Reclassify external appeals-committee decisions to source_kind='internal_committee'.
+
+    Identifies rows by source_type='appeals_committee' and updates source_kind + district.
+    Existing precedent_chunks remain — no re-embedding needed.
+    """
+    pool = await db.get_pool()
+    async with pool.acquire() as conn:
+        rows = await conn.fetch(
+            """SELECT id, case_number, court
+               FROM case_law
+               WHERE source_kind = 'external_upload'
+                 AND source_type = 'appeals_committee'"""
+        )
+
+    results = {"total": len(rows), "updated": 0, "dry_run": dry_run}
+
+    if dry_run:
+        results["updated"] = len(rows)
+        results["preview"] = [
+            {"case_number": r["case_number"], "court": r["court"], "district": _district_from_court(r["court"] or "")}
+            for r in rows
+        ]
+        return results
+
+    async with pool.acquire() as conn:
+        for row in rows:
+            district = _district_from_court(row["court"] or "")
+            await conn.execute(
+                """UPDATE case_law
+                   SET source_kind = 'internal_committee',
+                       district = CASE WHEN $2 <> '' THEN $2 ELSE district END
+                   WHERE id = $1""",
+                row["id"], district,
+            )
+        results["updated"] = len(rows)
+
+    logger.info("Migrated %d external appeals-committee rows to internal_committee", len(rows))
+    return results
+
+
+async def enrich_migrated_entries(dry_run: bool = False) -> dict:
+    """One-time enrichment: run metadata extraction + halacha extraction on all
+    internal_committee entries that are waiting (halacha_status='pending',
+    metadata never requested).
+
+    Metadata extraction will:
+      - Fix case_number from the decision header text
+      - Fill case_name from the parties line
+      - Fill date if missing
+
+    Halacha extraction queues the LLM-based halacha extraction job.
+    """
+    from legal_mcp.services import precedent_metadata_extractor, db as _db
+
+    pool = await _db.get_pool()
+    async with pool.acquire() as conn:
+        rows = await conn.fetch(
+            """SELECT id, case_number
+               FROM case_law
+               WHERE source_kind = 'internal_committee'
+                 AND halacha_extraction_status = 'pending'
+                 AND metadata_extraction_requested_at IS NULL
+               ORDER BY created_at"""
+        )
+
+    results = {
+        "total": len(rows),
+        "metadata_updated": 0,
+        "halachot_queued": 0,
+        "failed": 0,
+        "dry_run": dry_run,
+    }
+
+    if dry_run:
+        return results
+
+    for row in rows:
+        case_law_id = row["id"]
+        try:
+            meta = await precedent_metadata_extractor.extract_and_apply(
+                case_law_id, overwrite_case_number=True
+            )
+            if meta.get("status") in ("completed", "no_changes"):
+                results["metadata_updated"] += 1
+                logger.info(
+                    "enrich_migrated: %s → fields=%s",
+                    row["case_number"], meta.get("fields"),
+                )
+        except Exception as e:
+            logger.error("enrich_migrated metadata failed for %s: %s", row["case_number"], e)
+            results["failed"] += 1
+            continue
+
+        try:
+            await _db.request_halacha_extraction(case_law_id)
+            results["halachot_queued"] += 1
+        except Exception as e:
+            logger.error("enrich_migrated halacha queue failed for %s: %s", row["case_number"], e)
+
+    return results
+
+
+async def search_internal(
+    query: str,
+    *,
+    practice_area: str = "",
+    appeal_subtype: str = "",
+    district: str = "",
+    chair_name: str = "",
+    limit: int = 10,
+    include_halachot: bool = True,
+) -> list[dict]:
+    """Semantic search over internal committee decisions."""
+    from legal_mcp.services import hybrid_search
+
+    if not query.strip():
+        return []
+    query_vec = await embeddings.embed_query(query)
+    return await hybrid_search.search_precedent_library_hybrid(
+        query=query,
+        query_text_embedding=query_vec,
+        limit=limit,
+        practice_area=practice_area,
+        appeal_subtype=appeal_subtype,
+        include_halachot=include_halachot,
+        source_kind="internal_committee",
+        district=district,
+        chair_name=chair_name,
+    )

mcp-server/src/legal_mcp/services/lessons.py

@@ -123,7 +123,7 @@ SUMMARY_STRATEGIES = {
 
 DISCUSSION_RULES: dict[str, list[str]] = {
     "universal": [
-        "פרק הדיון = אסה רציפה. אין כותרות משנה (H2/H3). מעברים רק עם ביטויי מעבר טקסטואליים.",
+        "פרק הדיון = מאסה רציפה. אין כותרות משנה (H2/H3). מעברים רק עם ביטויי מעבר טקסטואליים.",
         "חריג יחיד לכותרות משנה: נושאים נפרדים לחלוטין (למשל: הקלה בגובה + התייחסות לטענות נוספות).",
         "טווח אורך סעיפים: 20 עד 600+ מילים. סעיף עם ציטוט מקיף = בלוק אחד שלם, לא שבירה לסעיפים קצרים.",
     ],

mcp-server/src/legal_mcp/services/precedent_library.py

@@ -257,11 +257,19 @@ async def reextract_halachot(
         case_law_id = UUID(case_law_id)
 
     record = await db.get_case_law(case_law_id)
-    if not record or record.get("source_kind") != "external_upload":
+    if not record:
-        raise ValueError("precedent not found or not chair-uploaded")
+        raise ValueError("precedent not found")
+    # Was restricted to source_kind='external_upload'; opened 2026-05-06 so
+    # internal_committee rows can also be re-extracted when ingest produced
+    # bad data. See note in db.request_metadata_extraction.
 
     await progress("extracting_halachot", 50, "מחלץ הלכות מחדש")
     result = await halacha_extractor.extract(case_law_id)
+    # Clear the queue timestamp on completion so the UI badge / worker queue
+    # don't keep showing this row. The queue worker (process_pending_extractions)
+    # already does this; mirror it here so per-record extraction drains too.
+    if result.get("status") in ("completed", "no_halachot"):
+        await db.clear_extraction_request(case_law_id, kind="halacha")
     await progress(
         "completed",
         100,
@@ -402,11 +410,16 @@ async def reextract_metadata(
         case_law_id = UUID(case_law_id)
 
     record = await db.get_case_law(case_law_id)
-    if not record or record.get("source_kind") != "external_upload":
+    if not record:
-        raise ValueError("precedent not found or not chair-uploaded")
+        raise ValueError("precedent not found")
+    # See note in db.request_metadata_extraction — opened to all source kinds.
 
     await progress("extracting_metadata", 40, "מחלץ מטא-דאטה (תקציר, תגיות)")
     result = await precedent_metadata_extractor.extract_and_apply(case_law_id)
+    # Clear the queue timestamp so the UI / worker stop showing this row.
+    # See note in reextract_halachot.
+    if result.get("status") in ("completed", "no_changes"):
+        await db.clear_extraction_request(case_law_id, kind="metadata")
     fields = result.get("fields") or []
     msg = (
         f"מולאו {len(fields)} שדות: {', '.join(fields)}"
@@ -425,13 +438,14 @@ async def delete_precedent(case_law_id: UUID | str) -> bool:
 
 
 async def get_precedent(case_law_id: UUID | str) -> dict | None:
-    """Get a precedent with its halachot attached."""
+    """Get a precedent with its halachot and related cases attached."""
     if isinstance(case_law_id, str):
         case_law_id = UUID(case_law_id)
     record = await db.get_case_law(case_law_id)
     if not record:
         return None
     record["halachot"] = await db.list_halachot(case_law_id=case_law_id, limit=500)
+    record["related_cases"] = await db.get_case_law_relations(case_law_id)
     return record
 
 

mcp-server/src/legal_mcp/services/precedent_metadata_extractor.py

@@ -50,7 +50,8 @@ METADATA_EXTRACTION_PROMPT = """אתה מסייע משפטי בכיר. קרא א
   "decision_date_iso": "YYYY-MM-DD — תאריך מתן ההחלטה כפי שמופיע בטקסט (בכותרת או בחתימה הסופית). אם לא ניתן לזהות במדויק — מחרוזת ריקה.",
   "precedent_level": "אחד מ-4: 'עליון' / 'מנהלי' / 'ועדת_ערר_ארצית' / 'ועדת_ערר_מחוזית'. בחר לפי הערכאה שמסומנת בכותרת הפסק. אם לא ברור — מחרוזת ריקה.",
   "source_type": "אחד מ-2: 'court_ruling' (פסק דין של בית משפט — עליון/מנהלי) / 'appeals_committee' (החלטה של ועדת ערר). אם לא ברור — מחרוזת ריקה.",
-  "court": "שם הערכאה כפי שהוא מופיע בכותרת (למשל 'בית המשפט העליון', 'בית המשפט המחוזי בירושלים בשבתו כבית משפט לעניינים מנהליים', 'ועדת הערר לתכנון ובניה פיצויים והיטלי השבחה — מחוז ירושלים'). מחרוזת ריקה אם לא ניתן לזהות."
+  "court": "שם הערכאה כפי שהוא מופיע בכותרת (למשל 'בית המשפט העליון', 'בית המשפט המחוזי בירושלים בשבתו כבית משפט לעניינים מנהליים', 'ועדת הערר לתכנון ובניה פיצויים והיטלי השבחה — מחוז ירושלים'). מחרוזת ריקה אם לא ניתן לזהות.",
+  "case_number_clean": "מספר הערר/תיק כפי שמופיע בכותרת — רק הספרות והאלכסון, למשל '1062/24' או '8031/21'. ללא המילה 'ערר', ללא שם הצדדים, ללא סוגריים. אם יש כמה עררים מאוחדים — הרשום הראשון. מחרוזת ריקה אם לא ניתן לזהות."
 }
 
 ## כללי איכות
@@ -161,12 +162,15 @@ async def extract_metadata(case_law_id: UUID | str) -> dict:
             out["source_type"] = st
     if isinstance(result.get("court"), str):
         out["court"] = result["court"].strip()
+    if isinstance(result.get("case_number_clean"), str):
+        out["case_number_clean"] = result["case_number_clean"].strip()
     return out
 
 
 async def apply_to_record(
     case_law_id: UUID | str,
     suggested: dict,
+    overwrite_case_number: bool = False,
 ) -> dict:
     """Merge suggested metadata into the case_law row, filling ONLY empty fields.
 
@@ -178,6 +182,9 @@ async def apply_to_record(
     case_name has special handling: if the current case_name equals the
     case_number (a tell-tale sign of the upload form sending the long
     citation into both fields), treat it as empty and overwrite.
+
+    overwrite_case_number: when True, update case_number from case_number_clean
+      even if the field already has a value (used for one-time migration enrichment).
     """
     if isinstance(case_law_id, str):
         case_law_id = UUID(case_law_id)
@@ -216,7 +223,17 @@ async def apply_to_record(
             fields_to_update["key_quote"] = s
 
     cur_tags = record.get("subject_tags") or []
-    if not cur_tags:
+    # Treat character-by-character corruption as empty. Early ingest
+    # pipelines stored a JSON string (`'["היטל השבחה"]'`) into a TEXT[]
+    # column, which Postgres split into individual chars:
+    # `['[', '"', 'ה', 'י', 'ט', 'ל', ' ', 'ה', 'ש', ...]`. Detection:
+    # 3+ elements where every element is at most 2 chars (legitimate
+    # tags are multi-character Hebrew words like `היטל_השבחה`).
+    is_corrupt = (
+        len(cur_tags) >= 3
+        and all(isinstance(t, str) and len(t) <= 2 for t in cur_tags)
+    )
+    if not cur_tags or is_corrupt:
         sug_tags = suggested.get("subject_tags") or []
         if sug_tags:
             fields_to_update["subject_tags"] = sug_tags
@@ -250,6 +267,11 @@ async def apply_to_record(
         if c:
             fields_to_update["court"] = c
 
+    if overwrite_case_number:
+        cn = (suggested.get("case_number_clean") or "").strip()
+        if cn:
+            fields_to_update["case_number"] = cn
+
     if not fields_to_update:
         return {"updated": False, "fields": []}
 
@@ -257,12 +279,15 @@ async def apply_to_record(
     return {"updated": True, "fields": list(fields_to_update.keys())}
 
 
-async def extract_and_apply(case_law_id: UUID | str) -> dict:
+async def extract_and_apply(
+    case_law_id: UUID | str,
+    overwrite_case_number: bool = False,
+) -> dict:
     """Convenience wrapper: extract → merge into row → return summary."""
     suggested = await extract_metadata(case_law_id)
     if not suggested:
         return {"status": "no_metadata", "fields": []}
-    result = await apply_to_record(case_law_id, suggested)
+    result = await apply_to_record(case_law_id, suggested, overwrite_case_number=overwrite_case_number)
     return {
         "status": "completed" if result["updated"] else "no_changes",
         "fields": result["fields"],

mcp-server/src/legal_mcp/tools/cases.py

@@ -13,7 +13,7 @@ from uuid import UUID
 import httpx
 
 from legal_mcp import config
-from legal_mcp.services import audit, db, git_sync, practice_area as pa
+from legal_mcp.services import audit, db, extractor, git_sync, practice_area as pa
 
 logger = logging.getLogger(__name__)
 
@@ -370,3 +370,66 @@ async def case_delete(case_number: str, remove_files: bool = False) -> str:
             result["removed_files"] = True
 
     return json.dumps(result, ensure_ascii=False, indent=2)
+
+
+async def case_get_final_text(case_number: str, max_chars: int = 0) -> str:
+    """קליטת טקסט ההחלטה הסופית (`סופי-{case}.docx` בתיקיית exports).
+
+    בניגוד ל-`document_get_text` שעובד על שורות בטבלת `documents`,
+    הקובץ הסופי הוא רק קובץ בתיקייה (נוצר על ידי `api_mark_final`).
+    תומך בכל הפורמטים ש-extractor.extract_text מטפל בהם — מנסה
+    `.docx` תחילה, ואז `.pdf`, `.doc`, `.rtf`, `.txt`, `.md`.
+
+    Args:
+        case_number: מספר תיק הערר
+        max_chars: אם >0, חתוך את הטקסט המוחזר לאורך הזה. 0 = הכל.
+    """
+    case_dir = config.find_case_dir(case_number)
+    exports_dir = case_dir / "exports"
+    final_stem = f"סופי-{case_number}"
+
+    final_path = None
+    for ext in (".docx", ".pdf", ".doc", ".rtf", ".txt", ".md"):
+        candidate = exports_dir / f"{final_stem}{ext}"
+        if candidate.exists():
+            final_path = candidate
+            break
+
+    if final_path is None:
+        return json.dumps({
+            "status": "not_found",
+            "case_number": case_number,
+            "expected_path": str(exports_dir / f"{final_stem}.docx"),
+            "tried_extensions": [".docx", ".pdf", ".doc", ".rtf", ".txt", ".md"],
+            "hint": (
+                "ההחלטה הסופית עדיין לא סומנה כ'סופית' ב-UI. "
+                "דפנה צריכה ללחוץ 'סמן כסופי' על קובץ הטיוטה הנכון."
+            ),
+        }, ensure_ascii=False, indent=2)
+
+    try:
+        text, page_count, _ = await extractor.extract_text(str(final_path))
+    except Exception as e:
+        logger.exception("case_get_final_text: extraction failed for %s", case_number)
+        return json.dumps({
+            "status": "error",
+            "case_number": case_number,
+            "file_path": str(final_path),
+            "error": str(e),
+        }, ensure_ascii=False, indent=2)
+
+    text = text or ""
+    truncated = False
+    if max_chars > 0 and len(text) > max_chars:
+        text = text[:max_chars]
+        truncated = True
+
+    return json.dumps({
+        "status": "ok",
+        "case_number": case_number,
+        "file_path": str(final_path),
+        "text_length": len(text),
+        "page_count": page_count,
+        "truncated": truncated,
+        "text": text,
+    }, ensure_ascii=False, indent=2)

mcp-server/src/legal_mcp/tools/precedent_library.py

@@ -116,6 +116,54 @@ async def precedent_library_get(case_law_id: str) -> str:
     return _ok(record)
 
 
+async def precedent_link_cases(
+    case_law_id_a: str,
+    case_law_id_b: str,
+    relation_type: str = "same_case_chain",
+) -> str:
+    """קישור שתי פסיקות כקשורות זו לזו (דו-כיווני). idempotent.
+
+    Args:
+        case_law_id_a: UUID של פסיקה ראשונה.
+        case_law_id_b: UUID של פסיקה שנייה.
+        relation_type: same_case_chain | overruled_by | distinguished
+    """
+    try:
+        a = UUID(case_law_id_a)
+        b = UUID(case_law_id_b)
+    except ValueError:
+        return _err("case_law_id לא תקין")
+    rec_a = await db.get_case_law(a)
+    rec_b = await db.get_case_law(b)
+    if not rec_a:
+        return _err(f"פסיקה {case_law_id_a} לא נמצאה")
+    if not rec_b:
+        return _err(f"פסיקה {case_law_id_b} לא נמצאה")
+    await db.add_case_law_relation(a, b, relation_type)
+    return _ok({
+        "linked": True,
+        "relation_type": relation_type,
+        "a": {"id": case_law_id_a, "case_number": rec_a.get("case_number"), "court": rec_a.get("court")},
+        "b": {"id": case_law_id_b, "case_number": rec_b.get("case_number"), "court": rec_b.get("court")},
+    })
+
+
+async def precedent_unlink_cases(case_law_id_a: str, case_law_id_b: str) -> str:
+    """הסרת קישור בין שתי פסיקות (דו-כיווני).
+
+    Args:
+        case_law_id_a: UUID של פסיקה ראשונה.
+        case_law_id_b: UUID של פסיקה שנייה.
+    """
+    try:
+        a = UUID(case_law_id_a)
+        b = UUID(case_law_id_b)
+    except ValueError:
+        return _err("case_law_id לא תקין")
+    await db.remove_case_law_relation(a, b)
+    return _ok({"unlinked": True, "a": case_law_id_a, "b": case_law_id_b})
+
+
 async def precedent_library_delete(case_law_id: str) -> str:
     """מחיקת פסיקה מהקורפוס. cascade: chunks + halachot."""
     try:

mcp-server/src/legal_mcp/tools/search.py

@@ -179,3 +179,63 @@ async def find_similar_cases(
         })
 
     return json.dumps(formatted, ensure_ascii=False, indent=2)
+
+
+async def search_internal_decisions(
+    query: str,
+    practice_area: str = "",
+    appeal_subtype: str = "",
+    district: str = "",
+    chair_name: str = "",
+    limit: int = 10,
+    include_halachot: bool = True,
+) -> str:
+    """חיפוש בהחלטות ועדות ערר לתכנון ובנייה (כל המחוזות).
+
+    Args:
+        query: שאילתת חיפוש בעברית
+        practice_area: rishuy_uvniya / betterment_levy / compensation_197
+        appeal_subtype: סינון לפי תת-סוג ערר
+        district: מחוז — ירושלים / מרכז / תל אביב / צפון / דרום / ארצי. ריק = כל המחוזות
+        chair_name: שם יו"ר הוועדה לסינון. ריק = כל היו"רים
+        limit: מספר תוצאות מקסימלי
+        include_halachot: האם לכלול הלכות שחולצו
+    """
+    from legal_mcp.services import internal_decisions as int_svc
+
+    results = await int_svc.search_internal(
+        query,
+        practice_area=practice_area,
+        appeal_subtype=appeal_subtype,
+        district=district,
+        chair_name=chair_name,
+        limit=limit,
+        include_halachot=include_halachot,
+    )
+
+    if not results:
+        return "לא נמצאו החלטות ועדת ערר רלוונטיות."
+
+    formatted = []
+    for r in results:
+        entry = {
+            "score": round(float(r["score"]), 4),
+            "type": r.get("type", "passage"),
+            "case_number": r.get("case_number"),
+            "case_name": r.get("case_name"),
+            "court": r.get("court"),
+            "district": r.get("district"),
+            "chair_name": r.get("chair_name"),
+            "decision_date": r.get("decision_date"),
+        }
+        if r.get("type") == "halacha":
+            entry["rule"] = r.get("rule_statement")
+            entry["quote"] = r.get("supporting_quote")
+            entry["rule_type"] = r.get("rule_type")
+        else:
+            entry["content"] = r.get("content", "")
+            entry["section"] = r.get("section_type")
+            entry["page"] = r.get("page_number")
+        formatted.append(entry)
+
+    return json.dumps(formatted, ensure_ascii=False, indent=2)

mcp-server/src/legal_mcp/tools/workflow.py

@@ -3,10 +3,13 @@
 from __future__ import annotations
 
 import json
+import logging
 from uuid import UUID
 
 from legal_mcp.services import db
 
+logger = logging.getLogger(__name__)
+
 
 async def workflow_status(case_number: str) -> str:
     """סטטוס תהליך עבודה מלא לתיק - מסמכים, עיבוד, טיוטות.
@@ -315,10 +318,29 @@ async def ingest_final_version(
 
     try:
         result = await learning_loop.process_final_version(case_id, final_text)
-        return json.dumps(result, default=str, ensure_ascii=False, indent=2)
     except ValueError as e:
         return json.dumps({"status": "error", "message": str(e)}, ensure_ascii=False, indent=2)
 
+    # Auto-ingest into internal committee decisions corpus (best-effort).
+    try:
+        from legal_mcp.services import internal_decisions as int_svc
+        await int_svc.ingest_internal_decision(
+            case_number=case_number,
+            case_name=case.get("title", ""),
+            decision_date=case.get("decision_date"),
+            chair_name=case.get("chair_name", ""),
+            district="ירושלים",
+            practice_area=case.get("practice_area", ""),
+            appeal_subtype=case.get("appeal_subtype", ""),
+            text=final_text,
+        )
+        result["internal_corpus_ingested"] = True
+    except Exception as e:
+        logger.warning("ingest_final_version: internal corpus ingestion failed (non-fatal): %s", e)
+        result["internal_corpus_ingested"] = False
+
+    return json.dumps(result, default=str, ensure_ascii=False, indent=2)
+
 
 # ── Chair feedback tools ──────────────────────────────────────────
 

scripts/.archive/run_curator_deepseek_test.sh

@@ -0,0 +1,87 @@
+#!/usr/bin/env bash
+# One-off A/B test runner: runs the Knowledge Curator (Hermes) on CMP-78 using
+# DeepSeek V4-Pro instead of the default Sonnet 4.5 (via marcus/sonnet gateway).
+# Compare against CMP-80 which runs with the default config.
+set -euo pipefail
+
+PROFILE_HOME="/home/chaim/.hermes/profiles/curator-cmp-deepseek"
+PAPERCLIP_API_URL="http://localhost:3100/api"
+# CMP curator agent's Paperclip key (from Infisical: nautilus /legal-ai HERMES_CURATOR_CMP_PAPERCLIP_KEY)
+PAPERCLIP_API_KEY="pcp_c87edcf306d06fce13fac701bb6d747191d61dba5b51e903"
+PAPERCLIP_TASK_ID="beb745e5-7195-40c5-9ac0-e9682c2c5184"   # CMP-78
+PAPERCLIP_TASK_KEY="$PAPERCLIP_TASK_ID"
+PAPERCLIP_TASK_TITLE="[ערר 1130-25] סקירת ידע — Knowledge Curator (DeepSeek A/B test)"
+PAPERCLIP_RUN_ID="deepseek-ab-$(date +%s)"
+PAPERCLIP_WAKE_REASON="manual_deepseek_ab_test"
+
+# Rendered prompt — copy of the curator template with mustache variables resolved
+# manually for CMP-78. We also add a clear "[ניסוי DeepSeek V4-Pro]" prefix so
+# the resulting comment is distinguishable from the default-Sonnet run on CMP-80.
+read -r -d '' PROMPT <<'EOF' || true
+אתה מנהל ידע (Knowledge Curator) של ועדת הערר. נעור על תיק שדפנה סימנה כסופי.
+
+תיק: [ערר 1130-25] סקירת ידע — Knowledge Curator
+issue ID: beb745e5-7195-40c5-9ac0-e9682c2c5184
+run reason: manual_deepseek_ab_test
+
+**הקשר חשוב — ניסוי A/B:** זוהי ריצה ידנית באמצעות DeepSeek V4-Pro במקום ה-Sonnet הרגיל. כל ה-comment שתפרסם חייב להתחיל בכותרת `[ניסוי DeepSeek V4-Pro]` כדי שנוכל להבדיל מהריצה המקבילה ב-CMP-80 (שרצה עם Sonnet). אל תעיר סוכנים אחרים. אל תיצור issues חדשים. אל תפתח interaction.
+
+הוראות:
+דפנה סימנה את ההחלטה הסופית של תיק 1130-25 כסופית.
+קובץ סופי: `סופי-1130-25.docx`
+
+סקור את ההחלטה מול skills/decision/SKILL.md ו-docs/legal-decision-lessons.md.
+חפש 3-5 דפוסי סגנון/דיון שלא תועדו. כתוב comment בעברית, ניטרלי, ממוספר.
+
+# שלבי ביצוע
+
+## 1. קונטקסט
+- קרא את MEMORY.md שלך (memory tool) — מה כבר זיהית.
+- קרא `/home/chaim/legal-ai/skills/decision/SKILL.md` (file tool) — מה כבר תועד.
+
+## 2. נתונים
+- `mcp__legal-ai__case_get` עם case_number `1130-25` — מטא-דאטה.
+- `mcp__legal-ai__case_get_final_text` עם case_number `1130-25` — קרא את הטקסט המלא של ההחלטה הסופית.
+- אם רלוונטי: `mcp__legal-ai__search_decisions` להשוואה לחלטות קודמות.
+
+## 3. ניתוח
+חפש 3-5 דפוסים/פערים. לכל ממצא: מה ראיתי + מה זה אומר + הצעה ניסוחית מדויקת.
+
+## 4. כתוב comment הממצאים
+```bash
+curl -sS -X POST -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  -d "$(jq -n --arg b "$BODY" '{body:$b}')" \
+  "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID/comments"
+```
+
+פורמט ה-body:
+- שורה ראשונה: `[ניסוי DeepSeek V4-Pro]`
+- אחר כך פסקה אחת מבוא קצרה
+- אחר כך הממצאים ממוספרים
+
+## 5. סגור את ה-issue
+```bash
+curl -sS -X PATCH -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  -d '{"status":"done"}' "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID"
+```
+
+# כללים
+- אל תעדכן קבצים (skills/, lessons.py, DB) בעצמך. רק comment.
+- אל תיצור issues חדשים.
+- אל תעיר סוכנים אחרים.
+- אל תפתח interaction.
+- בעיה? comment קצר עם הסיבה + סגור (status=done).
+EOF
+
+export HERMES_HOME="$PROFILE_HOME"
+export PAPERCLIP_API_URL PAPERCLIP_API_KEY PAPERCLIP_TASK_ID PAPERCLIP_TASK_KEY \
+  PAPERCLIP_TASK_TITLE PAPERCLIP_RUN_ID PAPERCLIP_WAKE_REASON
+
+echo "=== DeepSeek V4-Pro Curator A/B test on CMP-78 ==="
+echo "HERMES_HOME=$HERMES_HOME"
+echo "TASK_ID=$PAPERCLIP_TASK_ID"
+echo "RUN_ID=$PAPERCLIP_RUN_ID"
+echo "Starting Hermes..."
+echo "---"
+
+hermes -z "$PROMPT" --yolo chat 2>&1

scripts/.archive/run_curator_deepseek_test_v2.sh

@@ -0,0 +1,116 @@
+#!/usr/bin/env bash
+# A/B test runner #2: DeepSeek V4-Pro on CMP-78 — WITH interaction step
+# (matching the full Sonnet baseline workflow on CMP-80, including ask_user_questions).
+set -euo pipefail
+
+PROFILE_HOME="/home/chaim/.hermes/profiles/curator-cmp-deepseek"
+PAPERCLIP_API_URL="http://localhost:3100/api"
+PAPERCLIP_API_KEY="pcp_c87edcf306d06fce13fac701bb6d747191d61dba5b51e903"
+PAPERCLIP_TASK_ID="beb745e5-7195-40c5-9ac0-e9682c2c5184"   # CMP-78
+PAPERCLIP_TASK_KEY="$PAPERCLIP_TASK_ID"
+PAPERCLIP_TASK_TITLE="[ערר 1130-25] סקירת ידע — DeepSeek V4-Pro test #2 (with interaction)"
+PAPERCLIP_RUN_ID="deepseek-ab2-$(date +%s)"
+PAPERCLIP_WAKE_REASON="manual_deepseek_ab_test_v2_with_interaction"
+
+read -r -d '' PROMPT <<'EOF' || true
+אתה מנהל ידע (Knowledge Curator) של ועדת הערר. נעור על תיק שדפנה סימנה כסופי.
+
+תיק: [ערר 1130-25] סקירת ידע — Knowledge Curator
+issue ID: beb745e5-7195-40c5-9ac0-e9682c2c5184
+run reason: manual_deepseek_ab_test_v2_with_interaction
+
+**הקשר חשוב — ניסוי A/B #2:** זוהי ריצה שנייה ידנית באמצעות DeepSeek V4-Pro, הפעם **עם interaction מלא** כדי להשוות הוגנת מול ריצת Sonnet ב-CMP-80. כל הפלטים שתפרסם חייבים להתחיל בכותרת `[ניסוי DeepSeek V4-Pro #2 — עם interaction]`. אל תעיר סוכנים אחרים. אל תיצור issues חדשים.
+
+הוראות:
+דפנה סימנה את ההחלטה הסופית של תיק 1130-25 כסופית.
+קובץ סופי: `סופי-1130-25.docx`
+
+סקור את ההחלטה מול skills/decision/SKILL.md ו-docs/legal-decision-lessons.md.
+חפש 3-5 דפוסי סגנון/דיון שלא תועדו. כתוב comment בעברית, ניטרלי, ממוספר.
+
+# שלבי ביצוע
+
+## 1. קונטקסט
+- קרא את MEMORY.md שלך (memory tool) — מה כבר זיהית.
+- קרא `/home/chaim/legal-ai/skills/decision/SKILL.md` (file tool) — מה כבר תועד.
+
+## 2. נתונים
+- `mcp__legal-ai__case_get` עם case_number `1130-25` — מטא-דאטה.
+- `mcp__legal-ai__case_get_final_text` עם case_number `1130-25` — קרא את הטקסט המלא של ההחלטה הסופית.
+
+## 3. ניתוח
+חפש 3-5 דפוסים/פערים. לכל ממצא: מה ראיתי + מה זה אומר + הצעה ניסוחית מדויקת.
+
+## 4. כתוב comment הממצאים
+```bash
+curl -sS -X POST -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  -d "$(jq -n --arg b "$BODY" '{body:$b}')" \
+  "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID/comments"
+```
+
+פורמט ה-body:
+- שורה ראשונה: `[ניסוי DeepSeek V4-Pro #2 — עם interaction]`
+- אחר כך פסקה אחת מבוא קצרה
+- אחר כך הממצאים ממוספרים
+
+## 5. פתח interaction מסוג ask_user_questions
+זה השלב שעבד את Sonnet הרבה זמן — בוא נראה כמה זמן יקח לך.
+
+```bash
+curl -sS -X POST -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID/interactions" \
+  -d '{
+    "kind": "ask_user_questions",
+    "idempotencyKey": "curator-deepseek-v2:'"$PAPERCLIP_TASK_ID"':select",
+    "title": "[DeepSeek] איזה ממצאים שווים עדכון?",
+    "continuationPolicy": "wake_assignee",
+    "payload": {
+      "version": 1,
+      "submitLabel": "אשר בחירה",
+      "questions": [{
+        "id": "findings_to_propose",
+        "prompt": "סמן את הממצאים שאני אכין כהצעת עדכון ל-style guide",
+        "selectionMode": "multi",
+        "options": [
+          {"id":"f1","label":"<מילוי לפי ממצא 1>","description":"<תקציר>"},
+          {"id":"f2","label":"<מילוי לפי ממצא 2>","description":"<תקציר>"}
+        ]
+      }]
+    }
+  }'
+```
+
+מלא את ה-options לפי הממצאים שלך — אופציה אחת לכל ממצא ממוספר.
+
+## 6. עדכן issue ל-status=in_review (לא done — ממתינים לבחירת חיים)
+```bash
+curl -sS -X PATCH -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  -d '{"status":"in_review"}' "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID"
+```
+
+# כללים
+- אל תעדכן קבצים (skills/, lessons.py, DB) בעצמך. רק comment + interaction.
+- אל תיצור issues חדשים.
+- אל תעיר סוכנים אחרים.
+- בעיה? comment קצר עם הסיבה + סגור (status=done).
+EOF
+
+export HERMES_HOME="$PROFILE_HOME"
+export PAPERCLIP_API_URL PAPERCLIP_API_KEY PAPERCLIP_TASK_ID PAPERCLIP_TASK_KEY \
+  PAPERCLIP_TASK_TITLE PAPERCLIP_RUN_ID PAPERCLIP_WAKE_REASON
+
+echo "=== DeepSeek V4-Pro #2 (with interaction) — CMP-78 ==="
+echo "HERMES_HOME=$HERMES_HOME"
+echo "TASK_ID=$PAPERCLIP_TASK_ID"
+echo "RUN_ID=$PAPERCLIP_RUN_ID"
+echo "Started: $(date -u '+%Y-%m-%d %H:%M:%S UTC')"
+echo "---"
+
+START_EPOCH=$(date +%s)
+hermes -z "$PROMPT" --yolo chat 2>&1
+END_EPOCH=$(date +%s)
+DURATION=$((END_EPOCH - START_EPOCH))
+echo ""
+echo "=== Run finished ==="
+echo "Ended: $(date -u '+%Y-%m-%d %H:%M:%S UTC')"
+echo "Duration: ${DURATION}s ($((DURATION/60))m $((DURATION%60))s)"

scripts/.archive/run_curator_sonnet_rerun.sh

@@ -0,0 +1,106 @@
+#!/usr/bin/env bash
+# A/B test #3: Sonnet 4.5 re-run on CMP-78 — same task as DeepSeek #2 but with Sonnet.
+# Goal: check if Sonnet is consistent across runs (esp. the case-outcome detection),
+# given that the original Sonnet baseline on CMP-80 misread the outcome as "דחייה"
+# while the actual result is "קבלה חלקית".
+set -euo pipefail
+
+PROFILE_HOME="/home/chaim/.hermes/profiles/curator-cmp"   # default Sonnet profile
+PAPERCLIP_API_URL="http://localhost:3100/api"
+PAPERCLIP_API_KEY="pcp_c87edcf306d06fce13fac701bb6d747191d61dba5b51e903"
+PAPERCLIP_TASK_ID="beb745e5-7195-40c5-9ac0-e9682c2c5184"   # CMP-78
+PAPERCLIP_TASK_KEY="$PAPERCLIP_TASK_ID"
+PAPERCLIP_TASK_TITLE="[ערר 1130-25] סקירת ידע — Sonnet rerun (consistency check)"
+PAPERCLIP_RUN_ID="sonnet-rerun-$(date +%s)"
+PAPERCLIP_WAKE_REASON="manual_sonnet_consistency_rerun"
+
+read -r -d '' PROMPT <<'EOF' || true
+אתה מנהל ידע (Knowledge Curator) של ועדת הערר. נעור על תיק שדפנה סימנה כסופי.
+
+תיק: [ערר 1130-25] סקירת ידע — Knowledge Curator
+issue ID: beb745e5-7195-40c5-9ac0-e9682c2c5184
+run reason: manual_sonnet_consistency_rerun
+
+**הקשר חשוב — ניסוי A/B #3:** זוהי ריצה חוזרת ידנית באמצעות Sonnet 4.5 (אותו מודל שהריץ ב-CMP-80) — בדיקת עקביות. כל הפלטים שתפרסם חייבים להתחיל בכותרת `[ניסוי Sonnet 4.5 — ריצה חוזרת על CMP-78]`. אל תעיר סוכנים אחרים. אל תיצור issues חדשים.
+
+הוראות:
+דפנה סימנה את ההחלטה הסופית של תיק 1130-25 כסופית.
+קובץ סופי: `סופי-1130-25.docx`
+
+סקור את ההחלטה מול skills/decision/SKILL.md ו-docs/legal-decision-lessons.md.
+חפש 3-5 דפוסי סגנון/דיון שלא תועדו. כתוב comment בעברית, ניטרלי, ממוספר.
+
+# שלבי ביצוע
+
+## 1. קונטקסט
+- קרא את MEMORY.md שלך (memory tool) — מה כבר זיהית.
+- קרא `/home/chaim/legal-ai/skills/decision/SKILL.md` (file tool) — מה כבר תועד.
+
+## 2. נתונים
+- `mcp__legal-ai__case_get` עם case_number `1130-25` — מטא-דאטה.
+- `mcp__legal-ai__case_get_final_text` עם case_number `1130-25` — קרא את הטקסט המלא של ההחלטה הסופית.
+
+**שים לב במיוחד**: זהה במדויק את **תוצאת ההחלטה** (קבלה / קבלה חלקית / דחייה) על סמך הטקסט עצמו, לא על סמך הנחות.
+
+## 3. ניתוח
+חפש 3-5 דפוסים/פערים. לכל ממצא: מה ראיתי + מה זה אומר + הצעה ניסוחית מדויקת.
+
+## 4. כתוב comment הממצאים
+```bash
+curl -sS -X POST -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  -d "$(jq -n --arg b "$BODY" '{body:$b}')" \
+  "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID/comments"
+```
+
+פורמט ה-body:
+- שורה ראשונה: `[ניסוי Sonnet 4.5 — ריצה חוזרת על CMP-78]`
+- שורה שנייה: `**תוצאת ההחלטה הזו: <קבלה / קבלה חלקית / דחייה>** — ציין מפורשות
+- אחר כך פסקה אחת מבוא קצרה
+- אחר כך הממצאים ממוספרים
+
+## 5. פתח interaction מסוג ask_user_questions
+זהה לפלואו של Sonnet באמת. אם תקבל "Agent run id required" — נסה כמה דרכים, ואם לא הולך, פרסם comment עם רשימת אופציות לבחירה.
+
+```bash
+curl -sS -X POST -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID/interactions" \
+  -d '{
+    "kind": "ask_user_questions",
+    "idempotencyKey": "curator-sonnet-rerun:'"$PAPERCLIP_TASK_ID"':select",
+    "title": "[Sonnet rerun] איזה ממצאים שווים עדכון?",
+    "continuationPolicy": "wake_assignee",
+    "payload": {"version": 1, "submitLabel": "אשר בחירה",
+      "questions": [{"id": "findings_to_propose", "prompt": "סמן ממצאים", "selectionMode": "multi", "options": []}]}}'
+```
+
+## 6. עדכן issue ל-status=in_review
+```bash
+curl -sS -X PATCH -H "Authorization: Bearer $PAPERCLIP_API_KEY" -H "Content-Type: application/json" \
+  -d '{"status":"in_review"}' "$PAPERCLIP_API_URL/issues/$PAPERCLIP_TASK_ID"
+```
+
+# כללים
+- אל תעדכן קבצים בעצמך. רק comment + interaction.
+- אל תיצור issues חדשים.
+- אל תעיר סוכנים אחרים.
+EOF
+
+export HERMES_HOME="$PROFILE_HOME"
+export PAPERCLIP_API_URL PAPERCLIP_API_KEY PAPERCLIP_TASK_ID PAPERCLIP_TASK_KEY \
+  PAPERCLIP_TASK_TITLE PAPERCLIP_RUN_ID PAPERCLIP_WAKE_REASON
+
+echo "=== Sonnet 4.5 rerun (consistency check) — CMP-78 ==="
+echo "HERMES_HOME=$HERMES_HOME"
+echo "TASK_ID=$PAPERCLIP_TASK_ID"
+echo "RUN_ID=$PAPERCLIP_RUN_ID"
+echo "Started: $(date -u '+%Y-%m-%d %H:%M:%S UTC')"
+echo "---"
+
+START_EPOCH=$(date +%s)
+hermes -z "$PROMPT" --yolo chat 2>&1
+END_EPOCH=$(date +%s)
+DURATION=$((END_EPOCH - START_EPOCH))
+echo ""
+echo "=== Run finished ==="
+echo "Ended: $(date -u '+%Y-%m-%d %H:%M:%S UTC')"
+echo "Duration: ${DURATION}s ($((DURATION/60))m $((DURATION%60))s)"

scripts/SCRIPTS.md

@@ -10,7 +10,8 @@
 |--------|------|---------|-----------|
 | `pc.sh` | bash | **wrapper לכל קריאות Paperclip API מסוכנים** — מוסיף Authorization, X-Paperclip-Run-Id (audit trail), Content-Type, base URL. תחביר: `pc.sh <METHOD> <PATH> [BODY_JSON]`. אסור `curl` ישיר ל-`$PAPERCLIP_API_URL`. ראה `HEARTBEAT.md §0`. counterpart ב-Python: `web/paperclip_api.py`. | נקרא ע"י סוכנים |
 | `sync_missing_agent_skills.py` | python | סקריפט "אל-כשל" להוספת `paperclipSkillSync` ל-`הגהת מסמכים` ו-`מנתח משפטי` שפיספסו את ה-sync ההיסטורי (Gap #28). תומך `--verify`/`--dry-run`/`--apply`. גיבוי אוטומטי ל-`agents-pre-skill-sync-*.sql`. דורש `PAPERCLIP_BOARD_API_KEY` (Infisical /paperclip ב-nautilus env). idempotent. | חד-פעמי (בוצע 2026-05-04). שמור לרפרנס |
-| `sync_agents_across_companies.py` | python | **סנכרון סוכנים מ-CMP (1xxx, master) ל-CMPA (8xxx, mirror)** — Gap #25. משווה adapter_config (model/timeout/instructions/skills/etc), runtime_config (heartbeat), ושדות top-level (budget/metadata/icon/title/role). מסנן אוטומטית local skills שלא קיימים ב-mirror. לוגיקת subset (mirror יכול להחזיק יותר skills כי ה-API מוסיף required runtime skills). תומך `--verify`/`--dry-run`/`--apply [--only NAME]`. גיבוי אוטומטי. דורש `PAPERCLIP_BOARD_API_KEY`. **להריץ אחרי כל שינוי הגדרות ב-CMP.** | ידני אחרי כל שינוי |
+| `sync_agents_across_companies.py` | python | **סנכרון סוכנים מ-CMP (1xxx, master) ל-CMPA (8xxx, mirror)** — Gap #25. משווה adapter_config (model/timeout/instructions/skills/etc), runtime_config (heartbeat), ושדות top-level (budget/metadata/icon/title/role). מסנן אוטומטית local skills שלא קיימים ב-mirror. לוגיקת subset (mirror יכול להחזיק יותר skills כי ה-API מוסיף required runtime skills). תומך `--verify`/`--dry-run`/`--apply [--only NAME]`. גיבוי אוטומטי. דורש `PAPERCLIP_BOARD_API_KEY`. **להריץ אחרי כל שינוי הגדרות ב-CMP.** **⚠ אם `adapter_type` שונה בין CMP ל-CMPA — הסקריפט מדלג על הסוכן עם warning. בעת מעבר adapter (למשל ל-`deepseek_local`) חובה לעדכן ידנית בשתי החברות לפני sync.** | ידני אחרי כל שינוי |
+| `fix_paperclipai_skills_drift.py` | python | סקריפט חד-פעמי (בוצע 2026-05-04) שניקה drift על `paperclipai/*` skills בין CMP ל-CMPA. הסיר `paperclip-dev` מכל 14 הסוכנים, ודאג ש-`paperclip-converting-plans-to-tasks` קיים רק על CEO ו-analyst. תומך `--apply` (ברירת מחדל: dry-run). דורש `PAPERCLIP_BOARD_API_KEY`. נשמר לרפרנס למקרה שhdrift חוזר. | חד-פעמי (בוצע) |
 | `auto-sync-cases.sh` | bash | סנכרון תיקי ערר ל-Gitea — רץ כל דקה | `* * * * *` (cron) |
 | `backup-db.sh` | bash | גיבוי PostgreSQL יומי ל-`data/backups/` (gzip) | לתזמן: `0 2 * * *` |
 | `restore-db.sh` | bash | שחזור DB מגיבוי (companion ל-backup-db.sh) | ידני |
@@ -53,6 +54,9 @@
 | `seed-appeals.py` | seeding תיקי ערר ראשוניים ל-DB | MCP: `case_create()` |
 | `seed-knowledge.py` | seeding לקחים, ביטויי מעבר, פסיקה | MCP: `record_chair_feedback()`, `precedent_attach()` |
 | `validate-decision.py` | ולידציה מול block-schema | MCP: `validate_decision()` + `qa_validator.py` |
+| `run_curator_deepseek_test.sh` | A/B test #1 (2026-05-05) — Hermes Curator על CMP-78 דרך DeepSeek V4-Pro ב-`provider:custom`, ללא interaction. תוצאה: 6:33 דק׳, 5 ממצאי סגנון/לקסיקון, פי 3 מהיר מ-Sonnet baseline (CMP-80) ופי ~20 זול. **הסקריפט נקודתי לתיק 1130-25 — לא להריץ שוב** | החלפת Curator לאדפטר DeepSeek מקומי (בתהליך) |
+| `run_curator_deepseek_test_v2.sh` | A/B test #2 (2026-05-05) — אותו run אבל עם interaction. תוצאה: 9:08 דק׳, 5 ממצאים, היחיד מ-4 הריצות שזיהה תוצאה עובדתית נכונה (קבלה חלקית). interaction נכשל ב-API ("Agent run id required" בריצה ידנית). | החלפת Curator לאדפטר DeepSeek מקומי |
+| `run_curator_sonnet_rerun.sh` | A/B test #3 (2026-05-05) — ריצה חוזרת של Sonnet 4.5 על אותו CMP-78. תוצאה: 12:52 דק׳ (לעומת 20:13 בריצה המקורית — כי בלי לולאת interaction.json). זיהה תוצאה שגויה ("דחייה") **בעקביות עם הריצה המקורית** — Sonnet עקבי-בטעות, DeepSeek אקראי. | בדיקה חד-פעמית — לא להריץ שוב |
 
 ## סקריפטים שנמחקו (git history בלבד)
 

scripts/fix_paperclipai_skills_drift.py

@@ -0,0 +1,134 @@
+#!/usr/bin/env python3
+"""Fix paperclipai/* skill drift across CMP+CMPA agents.
+
+Goal: zero drift on paperclipai/* skills between master(CMP) and mirror(CMPA).
+
+Rules:
+* Remove ``paperclipai/paperclip/paperclip-dev`` from all 14 agents (not relevant
+  for legal work — it's for maintaining Paperclip itself).
+* Ensure ``paperclipai/paperclip/paperclip-converting-plans-to-tasks`` exists
+  on CEO + analyst agents in both companies (planning skill).
+* Remove ``paperclipai/paperclip/paperclip-converting-plans-to-tasks`` from any
+  other agent in either company that currently has it.
+
+Local/* and company/* skills are not touched — they're scoped to a company
+by design and drift is expected.
+
+Usage::
+
+    PAPERCLIP_BOARD_API_KEY=pbk_... python scripts/fix_paperclipai_skills_drift.py            # dry-run
+    PAPERCLIP_BOARD_API_KEY=pbk_... python scripts/fix_paperclipai_skills_drift.py --apply    # commit
+"""
+
+from __future__ import annotations
+
+import argparse
+import asyncio
+import os
+import sys
+
+import httpx
+
+PAPERCLIP_API_URL = os.environ.get("PAPERCLIP_API_URL", "http://localhost:3100")
+PAPERCLIP_BOARD_API_KEY = os.environ.get("PAPERCLIP_BOARD_API_KEY")
+
+COMPANIES = {
+    "licensing": ("CMP ", "42a7acd0-30c5-4cbd-ac97-7424f65df294"),
+    "betterment": ("CMPA", "8639e837-4c9d-47fa-a76b-95788d651896"),
+}
+
+DEV_SKILL = "paperclipai/paperclip/paperclip-dev"
+CONVERTING_SKILL = "paperclipai/paperclip/paperclip-converting-plans-to-tasks"
+
+# Hebrew names of the agents that should retain converting-plans-to-tasks.
+CONVERTING_TARGETS = {"עוזר משפטי", "מנתח משפטי"}
+
+
+def headers() -> dict[str, str]:
+    if not PAPERCLIP_BOARD_API_KEY:
+        sys.exit("PAPERCLIP_BOARD_API_KEY not set — fetch from Infisical first.")
+    return {
+        "Authorization": f"Bearer {PAPERCLIP_BOARD_API_KEY}",
+        "Content-Type": "application/json",
+    }
+
+
+async def fetch_company_agents(client: httpx.AsyncClient, company_id: str) -> list[dict]:
+    r = await client.get(f"{PAPERCLIP_API_URL}/api/companies/{company_id}/agents", headers=headers())
+    r.raise_for_status()
+    return r.json()
+
+
+def compute_changes(agent: dict) -> tuple[bool, list[str], list[str]]:
+    skill_sync = (agent.get("adapterConfig") or {}).get("paperclipSkillSync") or {}
+    old = list(skill_sync.get("desiredSkills") or [])
+    new = [s for s in old if s != DEV_SKILL]
+    if agent["name"] in CONVERTING_TARGETS:
+        if CONVERTING_SKILL not in new:
+            new.append(CONVERTING_SKILL)
+    else:
+        new = [s for s in new if s != CONVERTING_SKILL]
+    return (sorted(old) != sorted(new), old, new)
+
+
+async def patch_agent(
+    client: httpx.AsyncClient, agent_id: str, current_skill_sync: dict, new_skills: list[str]
+) -> None:
+    body = {
+        "adapterConfig": {
+            "paperclipSkillSync": {**current_skill_sync, "desiredSkills": new_skills},
+        }
+    }
+    r = await client.patch(
+        f"{PAPERCLIP_API_URL}/api/agents/{agent_id}", headers=headers(), json=body, timeout=15
+    )
+    r.raise_for_status()
+
+
+async def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--apply", action="store_true", help="commit changes (default: dry-run)")
+    args = parser.parse_args()
+
+    mode = "APPLY" if args.apply else "DRY-RUN"
+    print(f"=== {mode}: fixing paperclipai/* skill drift ===\n")
+
+    async with httpx.AsyncClient(timeout=15) as client:
+        all_agents: list[dict] = []
+        for label, (_, cid) in COMPANIES.items():
+            agents = await fetch_company_agents(client, cid)
+            for a in agents:
+                a["_company_label"] = COMPANIES[label][0]
+            all_agents.extend(agents)
+
+        changes_planned = 0
+        for a in sorted(all_agents, key=lambda x: (x["_company_label"], x["name"])):
+            changed, old, new = compute_changes(a)
+            label = a["_company_label"]
+            if not changed:
+                print(f"  {label}  {a['name']:20}  no change")
+                continue
+            changes_planned += 1
+            removed = sorted(set(old) - set(new))
+            added = sorted(set(new) - set(old))
+            print(f"  {label}  {a['name']:20}  -{len(removed)} +{len(added)}")
+            for s in removed:
+                print(f"    - {s}")
+            for s in added:
+                print(f"    + {s}")
+            if args.apply:
+                skill_sync = (a.get("adapterConfig") or {}).get("paperclipSkillSync") or {}
+                try:
+                    await patch_agent(client, a["id"], skill_sync, new)
+                    print("    ✓ patched")
+                except httpx.HTTPStatusError as e:
+                    print(f"    ✗ failed: {e.response.status_code} {e.response.text[:200]}")
+                    raise
+
+        print(f"\n{mode}: {changes_planned} agents would change")
+        if not args.apply and changes_planned > 0:
+            print("Run with --apply to commit.")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

skills/decision/SKILL.md

@@ -293,18 +293,25 @@ description: This skill should be used when writing legal decisions (החלטו
 
 ### 7.5 שלושה מקורות פסיקה — אל תבלבל
 
-המערכת מפרידה בין שלושה קורפוסי פסיקה. כל אחד מהם משמש למטרה אחרת ויש כלי MCP נפרד לחיפוש בו:
+המערכת מפרידה בין **ארבעה** קורפוסי פסיקה. כל אחד מהם משמש למטרה אחרת ויש כלי MCP נפרד לחיפוש בו:
 
 | קורפוס | טבלה | כלי חיפוש | תפקיד |
 |---|---|---|---|
-| תקדימי דפנה (סגנון) | `style_corpus` + `paragraph_embeddings` | `search_decisions` | החלטות שדפנה עצמה כתבה. מקור לסגנון, ניסוחים, ג'וריספרודנציה אישית. |
+| לימוד סגנון | `style_corpus` | (לא לחיפוש תוכן) | ממשק /training — ניתוח "הקול" של היו"ר: טון, ביטויי מעבר, מבנה פסקאות. **אין לחפש כאן תוכן משפטי.** |
-| ספריית הפסיקה הסמכותית | `case_law` (`source_kind='external_upload'`) + `halachot` | `search_precedent_library` | פסיקה חיצונית מחייבת — עליון, מנהלי, ועדות ערר אחרות — עם הלכות שאושרו ע"י דפנה. **המקור היחיד לציטוטים בבלוק י לפי CREAC.** |
+| החלטות ועדות ערר | `case_law` (`source_kind='internal_committee'`) + `halachot` | `search_internal_decisions` | **כל** ועדות הערר לתכנון ובנייה (כל המחוזות). מסונן לפי `district` ו-`chair_name`. מקור לעקביות פנימית ופרקטיקה ארצית. |
-| ציטוטים שצורפו ידנית | `case_precedents` | `precedent_search_library` | quotes שדפנה צירפה לתיק ספציפי בעבר. דומה לקורפוס סמכותי אך פר-תיק, ידני, לא עוברת חילוץ הלכות. |
+| פסיקת בתי משפט | `case_law` (`source_kind='external_upload'`) + `halachot` | `search_precedent_library` | בתי משפט: עליון, מנהלי, בג"ץ. **המקור היחיד לציטוטים מחייבים בבלוק י לפי CREAC.** |
+| ציטוטים ידניים | `case_precedents` | `precedent_search_library` | quotes שצורפו לתיק ספציפי בעבר. פר-תיק, ידני. |
 
-**הזרימה הסטנדרטית בבלוק י:**
+**הזרימה הסטנדרטית בבלוק י — חפש במקביל:**
-1. `search_decisions` קודם — בדוק אם דפנה כבר הכריעה בסוגיה דומה (חיסכון דוקטרינרי / הבחנה).
+1. `search_internal_decisions(district="ירושלים")` — האם ועדת ערר ירושלים הכריעה בסוגיה? (עקביות פנימית)
-2. `search_precedent_library` — חפש את הכלל המחייב והציטוט התומך לפסקת CREAC.
+   - אם יש תוצאה רלוונטית: הצג תחת **"החלטות ועדת ערר ירושלים"** והתייחס לה בניתוח.
-3. אם הצדדים הפנו לפסיקה שלא בקורפוס — דפנה מעלה אותה דרך `/precedents` ב-UI; חילוץ ההלכות אוטומטי וההלכות מחכות לאישורה.
+2. `search_internal_decisions()` (ריק = כל המחוזות) — פרקטיקה ארצית של ועדות אחרות.
+   - הצג תחת **"החלטות ועדות ערר אחרות"** — כמשל/השוואה, לא כמחייב.
+3. `search_precedent_library` — כלל מחייב מבית משפט לפסקת CREAC.
+   - הצג תחת **"פסיקת בתי משפט"** — זה המקור לציטוט מחייב.
+4. אם הצדדים הפנו לפסיקה שלא בקורפוס — דפנה מעלה אותה דרך `/precedents` ב-UI.
+
+**חשוב:** החלטות ועדת ערר הן פרקטיקה, לא מחייב. ציטוט מחייב בבלוק י מגיע רק מ-`search_precedent_library`.
 
 **איסור על המצאת ציטוטים** — ציטוט פסיקה חייב להגיע מאחד מהקורפוסים. אם אין הלכה מאושרת תומכת בנקודה — אל תמציא; ציין שהנושא דורש הוספת פסיקה לקורפוס.
 

web-ui/src/app/precedents/[id]/page.tsx

@@ -11,6 +11,7 @@ import { Skeleton } from "@/components/ui/skeleton";
 import { usePrecedent } from "@/lib/api/precedent-library";
 import { PrecedentEditSheet } from "@/components/precedents/precedent-edit-sheet";
 import { ExtractedHalachotSection } from "@/components/precedents/extracted-halachot";
+import { RelatedCasesSection } from "@/components/precedents/link-related-dialog";
 
 const PRACTICE_AREA_LABELS: Record<string, string> = {
   rishuy_uvniya: "רישוי ובנייה",
@@ -152,6 +153,15 @@ export default function PrecedentDetailPage({
               </CardContent>
             </Card>
 
+            <Card className="bg-surface border-rule shadow-sm">
+              <CardContent className="px-6 py-5">
+                <RelatedCasesSection
+                  caseId={id}
+                  related={data.related_cases ?? []}
+                />
+              </CardContent>
+            </Card>
+
             <Card className="bg-surface border-rule shadow-sm">
               <CardContent className="px-6 py-5">
                 <ExtractedHalachotSection halachot={data.halachot ?? []} />

web-ui/src/app/settings/_components/agents-tab.tsx

@@ -336,6 +336,7 @@ export function AgentsTab() {
     <div className="space-y-4">
       <Card className="bg-surface border-rule shadow-sm">
         <CardContent className="px-5 py-4 flex items-center justify-between gap-3 flex-wrap">
+          <div className="space-y-1">
             <div className="text-[0.85rem] text-ink-muted">
               {data.pairs.length} סוכנים × 2 חברות (CMP master / CMPA mirror)
               {totalDrift > 0 && (
@@ -347,6 +348,10 @@ export function AgentsTab() {
                 <span className="text-danger ms-2">· {missingCount} זוגות לא שלמים</span>
               )}
             </div>
+            <div className="text-[0.7rem] text-ink-light">
+              פערי skills מחושבים על paperclipai/* בלבד. local/* ו-company/* מסוננים — שם שונה בין החברות הוא צפוי.
+            </div>
+          </div>
           <Button
             variant="outline"
             size="sm"

web-ui/src/components/precedents/library-list-panel.tsx

@@ -1,7 +1,8 @@
 "use client";
 
 import { useState } from "react";
-import { Trash2, Plus, Pencil } from "lucide-react";
+import Link from "next/link";
+import { Trash2, Plus, Pencil, Wand2 } from "lucide-react";
 import { toast } from "sonner";
 import {
   Table, TableBody, TableCell, TableHead, TableHeader, TableRow,
@@ -16,11 +17,12 @@ import {
 import {
   usePrecedents,
   useDeletePrecedent,
+  useRequestHalachotExtraction,
   isPrecedentActive,
   type Precedent,
   type PracticeArea,
 } from "@/lib/api/precedent-library";
-import { PRACTICE_AREAS, PRECEDENT_LEVELS, practiceAreaShort } from "./practice-area";
+import { PRACTICE_AREAS, practiceAreaShort } from "./practice-area";
 import { PrecedentUploadSheet } from "./precedent-upload-sheet";
 import { PrecedentEditSheet } from "./precedent-edit-sheet";
 
@@ -33,17 +35,25 @@ function formatDate(iso: string | null) {
   }
 }
 
-/* The upload form (and Nevo PDFs) embed Unicode bidi marks (RTL/LTR/embedding/
- * isolate) inside the citation. They render as zero-width but visually push
- * the text away from the cell edge. Strip them for display only — DB still
- * has the original. */
 function cleanCitation(s: string | null | undefined): string {
   if (!s) return "—";
   return s.replace(/[‎‏‪-‮⁦-⁩]/g, "").trim();
 }
 
-/* Shimmering pill — used while extraction is actively running.
+// Show the "extract halachot" button only when the precedent hasn't had a
- * Visually distinct from the static "queued" / "completed" pills. */
+// successful (or even attempted) extraction yet. Hide while processing or
+// after completion to avoid duplicate requests.
+function needsHalachaExtraction(p: Precedent): boolean {
+  if (p.extraction_status !== "completed") return false; // text not ready
+  if (p.halacha_extraction_status === "processing") return false;
+  if (p.halacha_extraction_status === "completed") return false;
+  if (p.halacha_extraction_status === "pending" && p.halacha_extraction_requested_at) {
+    return false; // already queued
+  }
+  // Remaining cases: pending+no-requested_at (never tried) or failed (retry).
+  return true;
+}
+
 function ActivePill({ label }: { label: string }) {
   return (
     <Badge
@@ -55,12 +65,6 @@ function ActivePill({ label }: { label: string }) {
   );
 }
 
-/**
- * Five distinct states. The "queued" state is what the user actually
- * sees most of the time (after upload, both extractions are auto-queued
- * but the local MCP worker hasn't drained them yet); "מחלץ" / "מעבד"
- * shimmers and only appears while the extractor is actively running.
- */
 function StatusPill({ p }: { p: Precedent }) {
   if (p.extraction_status === "failed") {
     return (
@@ -103,37 +107,28 @@ function StatusPill({ p }: { p: Precedent }) {
       </Badge>
     );
   }
-  // halacha_extraction_status === "completed"
   if (p.halachot_count === 0) {
     return <Badge variant="outline">ללא הלכות</Badge>;
   }
   return (
-    <Badge
+    <Badge variant="outline" className="bg-gold-wash text-gold-deep border-gold/40">
-      variant="outline"
-      className="bg-gold-wash text-gold-deep border-gold/40"
-    >
       {p.approved_count}/{p.halachot_count} מאושרות
     </Badge>
   );
 }
 
-function PrecedentRow({
+function CourtRow({ p, onEdit }: { p: Precedent; onEdit: (id: string) => void }) {
-  p, onEdit,
-}: {
-  p: Precedent;
-  onEdit: (id: string) => void;
-}) {
   const del = useDeletePrecedent();
+  const reqHalachot = useRequestHalachotExtraction();
   const active = isPrecedentActive(p);
+  const showExtractHalachot = needsHalachaExtraction(p);
 
   const onDelete = async () => {
     if (active) {
-      toast.error(
+      toast.error("מתבצע עיבוד — לא ניתן למחוק עכשיו.");
-        "מתבצע עיבוד — לא ניתן למחוק עכשיו. המתיני לסיום או רעני את הדף.",
-      );
       return;
     }
-    if (!window.confirm(`למחוק את ${p.case_number}? cascade ימחק את ה-chunks וההלכות.`)) return;
+    if (!window.confirm(`למחוק את ${p.case_number}?`)) return;
     try {
       await del.mutateAsync(p.id);
       toast.success("נמחק");
@@ -142,65 +137,62 @@ function PrecedentRow({
     }
   };
 
+  const onExtractHalachot = async () => {
+    try {
+      await reqHalachot.mutateAsync(p.id);
+      toast.success("סומן לחילוץ הלכות. הריצי מ-Claude Code: precedent_process_pending_halachot");
+    } catch (e) {
+      toast.error(e instanceof Error ? e.message : "שגיאה");
+    }
+  };
+
   return (
     <TableRow className="border-rule hover:bg-gold-wash/30 align-top">
       <TableCell
-        /* shadcn TableCell defaults to whitespace-nowrap which forces the
-         * row wider than the container; for this column we want the long
-         * citation to wrap onto a second line instead of triggering the
-         * horizontal scroll on the table wrapper. min-w/max-w keeps the
-         * column wide enough to avoid awkward 2-word lines while leaving
-         * room for the other columns. */
         className="font-semibold text-navy text-right whitespace-normal break-words min-w-[280px] max-w-[420px] py-3"
         dir="rtl"
       >
-        <span dir="auto">{cleanCitation(p.case_number)}</span>
+        <Link href={`/precedents/${p.id}`} className="hover:underline hover:text-gold-deep" dir="auto">
+          {cleanCitation(p.case_number)}
+        </Link>
       </TableCell>
       <TableCell className="text-ink whitespace-normal break-words max-w-[260px] py-3">
         <div className="font-medium">{cleanCitation(p.case_name)}</div>
-        {p.court ? (
+        {p.court ? <div className="text-[0.72rem] text-ink-muted">{p.court}</div> : null}
-          <div className="text-[0.72rem] text-ink-muted">{p.court}</div>
-        ) : null}
-      </TableCell>
-      <TableCell className="text-ink-muted">
-        {p.date ? formatDate(p.date) : <span className="text-ink-light">—</span>}
       </TableCell>
+      <TableCell className="text-ink-muted">{p.date ? formatDate(p.date) : <span className="text-ink-light">—</span>}</TableCell>
       <TableCell>
         {p.practice_area ? (
           <Badge variant="outline" className="bg-navy-soft/40 text-navy border-navy/30">
             {practiceAreaShort(p.practice_area)}
           </Badge>
-        ) : (
+        ) : <span className="text-ink-light">—</span>}
-          <span className="text-ink-light">—</span>
-        )}
       </TableCell>
       <TableCell className="text-ink-muted text-[0.78rem]">
-        {p.precedent_level ? (
+        {p.precedent_level || <span className="text-ink-light">—</span>}
-          p.precedent_level
-        ) : (
-          <span className="text-ink-light">—</span>
-        )}
-      </TableCell>
-      <TableCell>
-        <StatusPill p={p} />
       </TableCell>
+      <TableCell><StatusPill p={p} /></TableCell>
       <TableCell className="text-end">
         <div className="flex items-center gap-1 justify-end">
-          <Button
+          <Button variant="ghost" size="sm" onClick={() => onEdit(p.id)}
-            variant="ghost" size="sm" onClick={() => onEdit(p.id)}
+            aria-label={`ערוך את ${p.case_number}`} title="ערוך"
-            aria-label={`ערוך את ${p.case_number}`}
+            className="text-ink-muted hover:text-navy">
-            title="ערוך פרטים"
-            className="text-ink-muted hover:text-navy"
-          >
             <Pencil className="w-4 h-4" />
           </Button>
-          <Button
+          {showExtractHalachot && (
-            variant="ghost" size="sm" onClick={onDelete}
+            <Button variant="ghost" size="sm" onClick={onExtractHalachot}
+              disabled={reqHalachot.isPending}
+              aria-label={`חלץ הלכות מ-${p.case_number}`}
+              title="חלץ הלכות"
+              className="text-gold-deep hover:text-gold-deep hover:bg-gold-wash disabled:opacity-30">
+              <Wand2 className="w-4 h-4" />
+            </Button>
+          )}
+          <Button variant="ghost" size="sm" onClick={onDelete}
             disabled={del.isPending || active}
             aria-label={`מחק את ${p.case_number}`}
             title={active ? "מתבצע עיבוד — לא ניתן למחוק" : "מחק"}
-            className="text-danger hover:text-danger hover:bg-danger-bg disabled:opacity-30"
+            className="text-danger hover:text-danger hover:bg-danger-bg disabled:opacity-30">
-          >
             <Trash2 className="w-4 h-4" />
           </Button>
         </div>
@@ -209,33 +201,134 @@ function PrecedentRow({
   );
 }
 
+function CommitteeRow({ p, onEdit }: { p: Precedent; onEdit: (id: string) => void }) {
+  const del = useDeletePrecedent();
+  const reqHalachot = useRequestHalachotExtraction();
+  const active = isPrecedentActive(p);
+  const showExtractHalachot = needsHalachaExtraction(p);
+
+  const onDelete = async () => {
+    if (active) {
+      toast.error("מתבצע עיבוד — לא ניתן למחוק עכשיו.");
+      return;
+    }
+    if (!window.confirm(`למחוק את ${p.case_number}?`)) return;
+    try {
+      await del.mutateAsync(p.id);
+      toast.success("נמחק");
+    } catch (e) {
+      toast.error(e instanceof Error ? e.message : "שגיאה");
+    }
+  };
+
+  const onExtractHalachot = async () => {
+    try {
+      await reqHalachot.mutateAsync(p.id);
+      toast.success("סומן לחילוץ הלכות. הריצי מ-Claude Code: precedent_process_pending_halachot");
+    } catch (e) {
+      toast.error(e instanceof Error ? e.message : "שגיאה");
+    }
+  };
+
+  return (
+    <TableRow className="border-rule hover:bg-gold-wash/30 align-top">
+      <TableCell
+        className="font-semibold text-navy text-right whitespace-normal break-words min-w-[200px] max-w-[320px] py-3"
+        dir="rtl"
+      >
+        <Link href={`/precedents/${p.id}`} className="hover:underline hover:text-gold-deep" dir="auto">
+          {cleanCitation(p.case_number)}
+        </Link>
+      </TableCell>
+      <TableCell className="text-ink whitespace-normal break-words max-w-[220px] py-3">
+        <div className="font-medium">{cleanCitation(p.case_name)}</div>
+      </TableCell>
+      <TableCell className="text-ink-muted text-[0.78rem]">
+        {p.district || <span className="text-ink-light">—</span>}
+      </TableCell>
+      <TableCell className="text-ink-muted text-[0.78rem]">
+        {p.chair_name || <span className="text-ink-light">—</span>}
+      </TableCell>
+      <TableCell className="text-ink-muted">{p.date ? formatDate(p.date) : <span className="text-ink-light">—</span>}</TableCell>
+      <TableCell>
+        {p.practice_area ? (
+          <Badge variant="outline" className="bg-navy-soft/40 text-navy border-navy/30">
+            {practiceAreaShort(p.practice_area)}
+          </Badge>
+        ) : <span className="text-ink-light">—</span>}
+      </TableCell>
+      <TableCell><StatusPill p={p} /></TableCell>
+      <TableCell className="text-end">
+        <div className="flex items-center gap-1 justify-end">
+          <Button variant="ghost" size="sm" onClick={() => onEdit(p.id)}
+            aria-label={`ערוך את ${p.case_number}`} title="ערוך"
+            className="text-ink-muted hover:text-navy">
+            <Pencil className="w-4 h-4" />
+          </Button>
+          {showExtractHalachot && (
+            <Button variant="ghost" size="sm" onClick={onExtractHalachot}
+              disabled={reqHalachot.isPending}
+              aria-label={`חלץ הלכות מ-${p.case_number}`}
+              title="חלץ הלכות"
+              className="text-gold-deep hover:text-gold-deep hover:bg-gold-wash disabled:opacity-30">
+              <Wand2 className="w-4 h-4" />
+            </Button>
+          )}
+          <Button variant="ghost" size="sm" onClick={onDelete}
+            disabled={del.isPending || active}
+            aria-label={`מחק את ${p.case_number}`}
+            title={active ? "מתבצע עיבוד — לא ניתן למחוק" : "מחק"}
+            className="text-danger hover:text-danger hover:bg-danger-bg disabled:opacity-30">
+            <Trash2 className="w-4 h-4" />
+          </Button>
+        </div>
+      </TableCell>
+    </TableRow>
+  );
+}
+
+function TableSkeleton({ cols }: { cols: number }) {
+  return (
+    <>
+      {[...Array(4)].map((_, i) => (
+        <TableRow key={i} className="border-rule">
+          {[...Array(cols)].map((_, j) => (
+            <TableCell key={j}><Skeleton className="h-5 w-full" /></TableCell>
+          ))}
+        </TableRow>
+      ))}
+    </>
+  );
+}
+
 export function LibraryListPanel() {
   const [practiceArea, setPracticeArea] = useState<PracticeArea>("");
-  const [precedentLevel, setPrecedentLevel] = useState("");
   const [search, setSearch] = useState("");
   const [uploadOpen, setUploadOpen] = useState(false);
   const [editingId, setEditingId] = useState<string | null>(null);
 
-  const { data, isPending, error } = usePrecedents({
+  const sharedFilters = {
     practiceArea: practiceArea || undefined,
-    precedentLevel: precedentLevel || undefined,
     search: search.trim() || undefined,
     limit: 200,
-  });
+  };
+
+  const courts = usePrecedents({ ...sharedFilters, sourceKind: "external_upload", sourceType: "court_ruling" });
+  const committee = usePrecedents({ ...sharedFilters, sourceKind: "all_committees" });
 
   return (
-    <div className="space-y-4">
+    <div className="space-y-8">
+      {/* Shared filters */}
       <div className="flex items-end gap-3 flex-wrap">
         <div className="flex-1 min-w-[200px]">
-          <label className="text-[0.78rem] text-ink-muted">חיפוש (מספר תיק / שם / תקציר)</label>
+          <label className="text-[0.78rem] text-ink-muted">חיפוש (מספר תיק / שם)</label>
           <Input
             value={search}
             onChange={(e) => setSearch(e.target.value)}
-            placeholder="עע&quot;מ 3975/22"
+            placeholder="עע״מ 3975/22 / 1200-25"
             dir="rtl"
           />
         </div>
-
         <div className="min-w-[180px]">
           <label className="text-[0.78rem] text-ink-muted">תחום</label>
           <Select value={practiceArea || "_all"} onValueChange={(v) => setPracticeArea(v === "_all" ? "" : v as PracticeArea)}>
@@ -248,29 +341,25 @@ export function LibraryListPanel() {
             </SelectContent>
           </Select>
         </div>
-
-        <div className="min-w-[170px]">
-          <label className="text-[0.78rem] text-ink-muted">רמת תקדים</label>
-          <Select value={precedentLevel || "_all"} onValueChange={(v) => setPrecedentLevel(v === "_all" ? "" : v)}>
-            <SelectTrigger><SelectValue placeholder="הכל" /></SelectTrigger>
-            <SelectContent>
-              <SelectItem value="_all">הכל</SelectItem>
-              {PRECEDENT_LEVELS.map((l) => (
-                <SelectItem key={l.value} value={l.value}>{l.label}</SelectItem>
-              ))}
-            </SelectContent>
-          </Select>
-        </div>
-
         <Button onClick={() => setUploadOpen(true)} className="bg-navy text-parchment hover:bg-navy-soft">
           <Plus className="w-4 h-4 me-1" />
           העלאת פסיקה
         </Button>
       </div>
 
-      {error ? (
+      {/* Table 1 — Court rulings */}
-        <div className="rounded bg-danger-bg border border-danger/40 px-6 py-5 text-danger text-center">
+      <section>
-          {error.message}
+        <div className="flex items-center gap-2 mb-3">
+          <h3 className="text-base font-semibold text-navy">פסיקת בתי משפט</h3>
+          {courts.data && (
+            <Badge variant="outline" className="text-ink-muted text-[0.72rem]">
+              {courts.data.count}
+            </Badge>
+          )}
+        </div>
+        {courts.error ? (
+          <div className="rounded bg-danger-bg border border-danger/40 px-6 py-4 text-danger text-center text-sm">
+            {courts.error.message}
           </div>
         ) : (
           <div className="rounded-lg border border-rule bg-surface shadow-sm overflow-hidden">
@@ -287,31 +376,73 @@ export function LibraryListPanel() {
                 </TableRow>
               </TableHeader>
               <TableBody>
-              {isPending ? (
+                {courts.isPending ? (
-                [...Array(5)].map((_, i) => (
+                  <TableSkeleton cols={7} />
-                  <TableRow key={i} className="border-rule">
+                ) : !courts.data?.items.length ? (
-                    {[...Array(7)].map((_, j) => (
-                      <TableCell key={j}>
-                        <Skeleton className="h-5 w-full" />
-                      </TableCell>
-                    ))}
-                  </TableRow>
-                ))
-              ) : !data?.items.length ? (
                   <TableRow className="border-rule">
-                  <TableCell colSpan={7} className="text-center text-ink-muted py-10">
+                    <TableCell colSpan={7} className="text-center text-ink-muted py-8">
-                    אין פסיקה בקורפוס. העלה את פסק הדין הראשון.
+                      אין פסיקת בתי משפט בקורפוס.
                     </TableCell>
                   </TableRow>
                 ) : (
-                data.items.map((p) => (
+                  courts.data.items.map((p) => (
-                  <PrecedentRow key={p.id} p={p} onEdit={setEditingId} />
+                    <CourtRow key={p.id} p={p} onEdit={setEditingId} />
                   ))
                 )}
               </TableBody>
             </Table>
           </div>
         )}
+      </section>
+
+      {/* Table 2 — Appeals committee decisions */}
+      <section>
+        <div className="flex items-center gap-2 mb-3">
+          <h3 className="text-base font-semibold text-navy">החלטות ועדות ערר</h3>
+          {committee.data && (
+            <Badge variant="outline" className="text-ink-muted text-[0.72rem]">
+              {committee.data.count}
+            </Badge>
+          )}
+        </div>
+        {committee.error ? (
+          <div className="rounded bg-danger-bg border border-danger/40 px-6 py-4 text-danger text-center text-sm">
+            {committee.error.message}
+          </div>
+        ) : (
+          <div className="rounded-lg border border-rule bg-surface shadow-sm overflow-hidden">
+            <Table>
+              <TableHeader className="bg-rule-soft/60">
+                <TableRow className="border-rule">
+                  <TableHead className="text-navy text-right">מספר ערר</TableHead>
+                  <TableHead className="text-navy text-right">שם</TableHead>
+                  <TableHead className="text-navy text-right">מחוז</TableHead>
+                  <TableHead className="text-navy text-right">יו״ר</TableHead>
+                  <TableHead className="text-navy text-right">תאריך</TableHead>
+                  <TableHead className="text-navy text-right">תחום</TableHead>
+                  <TableHead className="text-navy text-right">הלכות</TableHead>
+                  <TableHead className="text-navy" />
+                </TableRow>
+              </TableHeader>
+              <TableBody>
+                {committee.isPending ? (
+                  <TableSkeleton cols={8} />
+                ) : !committee.data?.items.length ? (
+                  <TableRow className="border-rule">
+                    <TableCell colSpan={8} className="text-center text-ink-muted py-8">
+                      אין החלטות ועדת ערר בקורפוס.
+                    </TableCell>
+                  </TableRow>
+                ) : (
+                  committee.data.items.map((p) => (
+                    <CommitteeRow key={p.id} p={p} onEdit={setEditingId} />
+                  ))
+                )}
+              </TableBody>
+            </Table>
+          </div>
+        )}
+      </section>
 
       <PrecedentUploadSheet open={uploadOpen} onOpenChange={setUploadOpen} />
       <PrecedentEditSheet

web-ui/src/components/precedents/link-related-dialog.tsx

@@ -0,0 +1,222 @@
+"use client";
+
+import { useState } from "react";
+import { Link2, Loader2, X } from "lucide-react";
+import { toast } from "sonner";
+import {
+  Dialog,
+  DialogContent,
+  DialogHeader,
+  DialogTitle,
+} from "@/components/ui/dialog";
+import { Button } from "@/components/ui/button";
+import { Input } from "@/components/ui/input";
+import { Badge } from "@/components/ui/badge";
+import {
+  usePrecedents,
+  useLinkRelatedCase,
+  useUnlinkRelatedCase,
+  RelatedCase,
+} from "@/lib/api/precedent-library";
+
+const LEVEL_LABELS: Record<string, string> = {
+  "עליון": "עליון",
+  "מנהלי": "מנהלי",
+  "ועדת_ערר_ארצית": "ארצי",
+  "ועדת_ערר_מחוזית": "מחוזי",
+};
+
+const LEVEL_COLORS: Record<string, string> = {
+  "עליון": "bg-red-50 text-red-700 border-red-200",
+  "מנהלי": "bg-orange-50 text-orange-700 border-orange-200",
+  "ועדת_ערר_ארצית": "bg-blue-50 text-blue-700 border-blue-200",
+  "ועדת_ערר_מחוזית": "bg-green-50 text-green-700 border-green-200",
+};
+
+// ── Search Dialog ────────────────────────────────────────────────────
+
+type DialogProps = {
+  caseId: string;
+  currentRelated: RelatedCase[];
+  open: boolean;
+  onOpenChange: (open: boolean) => void;
+};
+
+function LinkDialog({ caseId, currentRelated, open, onOpenChange }: DialogProps) {
+  const [query, setQuery] = useState("");
+  const { mutateAsync: linkCase, isPending } = useLinkRelatedCase(caseId);
+  const alreadyLinkedIds = new Set([...currentRelated.map((r) => r.id), caseId]);
+
+  const { data, isPending: searching } = usePrecedents(
+    query.length >= 2 ? { search: query, limit: 10 } : {},
+  );
+
+  const candidates = (data?.items ?? []).filter((p) => !alreadyLinkedIds.has(p.id));
+
+  async function handleLink(relatedId: string) {
+    try {
+      await linkCase({ relatedId });
+      toast.success("הפסיקות קושרו");
+      setQuery("");
+    } catch {
+      toast.error("שגיאה בקישור");
+    }
+  }
+
+  return (
+    <Dialog open={open} onOpenChange={onOpenChange}>
+      <DialogContent className="max-w-lg" dir="rtl">
+        <DialogHeader>
+          <DialogTitle className="text-navy">קשר החלטה קשורה</DialogTitle>
+        </DialogHeader>
+
+        <div className="space-y-3">
+          <Input
+            placeholder="חפש לפי מספר תיק, שם, ערכאה..."
+            value={query}
+            onChange={(e) => setQuery(e.target.value)}
+            autoFocus
+            dir="rtl"
+          />
+
+          {query.length >= 2 && (
+            <div className="space-y-1 max-h-72 overflow-y-auto">
+              {searching ? (
+                <div className="flex items-center gap-2 text-ink-muted text-sm py-3">
+                  <Loader2 className="w-3.5 h-3.5 animate-spin" /> מחפש...
+                </div>
+              ) : candidates.length === 0 ? (
+                <p className="text-ink-muted text-sm py-3 text-center">לא נמצאו תוצאות</p>
+              ) : (
+                candidates.map((p) => (
+                  <button
+                    key={p.id}
+                    onClick={() => handleLink(p.id)}
+                    disabled={isPending}
+                    className="w-full text-right flex items-center justify-between gap-3 px-3 py-2.5 rounded-lg border border-rule hover:bg-surface/60 transition-colors disabled:opacity-50"
+                  >
+                    <div className="min-w-0 flex-1">
+                      <div className="text-sm font-medium text-navy truncate">
+                        {p.case_name || p.case_number}
+                      </div>
+                      <div className="text-[0.72rem] text-ink-muted font-mono" dir="ltr">
+                        {p.case_number}
+                      </div>
+                    </div>
+                    {p.precedent_level && (
+                      <Badge
+                        variant="outline"
+                        className={`text-[0.62rem] shrink-0 ${LEVEL_COLORS[p.precedent_level] ?? ""}`}
+                      >
+                        {LEVEL_LABELS[p.precedent_level] ?? p.precedent_level}
+                      </Badge>
+                    )}
+                  </button>
+                ))
+              )}
+            </div>
+          )}
+
+          {query.length > 0 && query.length < 2 && (
+            <p className="text-ink-muted text-xs">הקלד לפחות 2 תווים</p>
+          )}
+        </div>
+      </DialogContent>
+    </Dialog>
+  );
+}
+
+// ── Related Case Card ────────────────────────────────────────────────
+
+function RelatedCaseCard({ caseId, related }: { caseId: string; related: RelatedCase }) {
+  const { mutateAsync: unlinkCase, isPending } = useUnlinkRelatedCase(caseId);
+
+  async function handleUnlink() {
+    try {
+      await unlinkCase(related.id);
+      toast.success("הקישור הוסר");
+    } catch {
+      toast.error("שגיאה בהסרת הקישור");
+    }
+  }
+
+  return (
+    <div className="flex items-center justify-between gap-3 px-3 py-2.5 rounded-lg border border-rule bg-surface">
+      <a
+        href={`/precedents/${related.id}`}
+        className="min-w-0 flex-1 hover:opacity-80 transition-opacity"
+      >
+        <div className="text-sm font-medium text-navy truncate">
+          {related.case_name || related.case_number}
+        </div>
+        <div className="flex items-center gap-2 mt-0.5 flex-wrap">
+          {related.precedent_level && (
+            <Badge
+              variant="outline"
+              className={`text-[0.62rem] ${LEVEL_COLORS[related.precedent_level] ?? ""}`}
+            >
+              {LEVEL_LABELS[related.precedent_level] ?? related.precedent_level}
+            </Badge>
+          )}
+          {related.court && (
+            <span className="text-[0.7rem] text-ink-muted truncate">{related.court}</span>
+          )}
+          {related.date && (
+            <span className="text-[0.7rem] text-ink-muted tabular-nums" dir="ltr">
+              {related.date.slice(0, 10)}
+            </span>
+          )}
+        </div>
+      </a>
+      <button
+        onClick={handleUnlink}
+        disabled={isPending}
+        className="p-1 rounded hover:bg-danger-bg hover:text-danger transition-colors text-ink-muted disabled:opacity-40 shrink-0"
+        title="הסר קישור"
+      >
+        <X className="w-3.5 h-3.5" />
+      </button>
+    </div>
+  );
+}
+
+// ── Public section component ─────────────────────────────────────────
+
+type SectionProps = {
+  caseId: string;
+  related: RelatedCase[];
+};
+
+export function RelatedCasesSection({ caseId, related }: SectionProps) {
+  const [dialogOpen, setDialogOpen] = useState(false);
+
+  return (
+    <div className="space-y-3">
+      <div className="flex items-center justify-between">
+        <h3 className="text-navy text-sm font-semibold">
+          החלטות קשורות{related.length > 0 ? ` (${related.length})` : ""}
+        </h3>
+        <Button variant="outline" size="sm" onClick={() => setDialogOpen(true)}>
+          <Link2 className="w-3.5 h-3.5 me-1" /> קשר החלטה
+        </Button>
+      </div>
+
+      {related.length === 0 ? (
+        <p className="text-ink-muted text-sm">אין החלטות קשורות עדיין</p>
+      ) : (
+        <div className="space-y-1.5">
+          {related.map((r) => (
+            <RelatedCaseCard key={r.id} caseId={caseId} related={r} />
+          ))}
+        </div>
+      )}
+
+      <LinkDialog
+        caseId={caseId}
+        currentRelated={related}
+        open={dialogOpen}
+        onOpenChange={setDialogOpen}
+      />
+    </div>
+  );
+}

web-ui/src/components/precedents/practice-area.ts

@@ -35,3 +35,17 @@ export function practiceAreaShort(value: string | null | undefined): string {
   const match = PRACTICE_AREAS.find((p) => p.value === value);
   return match ? match.short : value;
 }
+
+/**
+ * Display label for ``appeal_subtype``. The field is free text — the chair
+ * usually types specific subjects like "תכנית רחביה" or "סופיות ההחלטה" —
+ * but the LLM extractor sometimes mistakenly stuffs the practice_area
+ * enum value (``betterment_levy``, ``rishuy_uvniya``, ``compensation_197``)
+ * in there. When that happens, translate to Hebrew so the UI doesn't
+ * leak English. Otherwise return the value unchanged.
+ */
+export function appealSubtypeLabel(value: string | null | undefined): string {
+  if (!value) return "";
+  const enumMatch = PRACTICE_AREAS.find((p) => p.value === value);
+  return enumMatch ? enumMatch.label : value;
+}

web-ui/src/components/precedents/precedent-edit-sheet.tsx

@@ -1,6 +1,6 @@
 "use client";
 
-import { useEffect, useState } from "react";
+import { useState } from "react";
 import { Save, Sparkles } from "lucide-react";
 import { toast } from "sonner";
 import {
@@ -22,7 +22,7 @@ import {
   type SourceType,
 } from "@/lib/api/precedent-library";
 import {
-  PRACTICE_AREAS, PRECEDENT_LEVELS, SOURCE_TYPES,
+  PRACTICE_AREAS, PRECEDENT_LEVELS, SOURCE_TYPES, appealSubtypeLabel,
 } from "./practice-area";
 import { ExtractedHalachotSection } from "./extracted-halachot";
 
@@ -65,17 +65,19 @@ export function PrecedentEditSheet({ caseLawId, onOpenChange }: Props) {
 
   const [form, setForm] = useState<FormState>(EMPTY);
 
-  // Hydrate form when the record loads.
+  // React-approved derived-state pattern: sync form whenever a different
-  useEffect(() => {
+  // record arrives (including after save+refetch). Using setState during
-    if (!record) return;
+  // render avoids the one-frame flash that useEffect would produce.
-    // eslint-disable-next-line react-hooks/set-state-in-effect
+  const [syncedRecordId, setSyncedRecordId] = useState<string | null>(null);
+  if (record && record.id !== syncedRecordId) {
+    setSyncedRecordId(record.id as string);
     setForm({
       citation: record.case_number || "",
       case_name: record.case_name || "",
       court: record.court || "",
       decision_date: record.date ? record.date.slice(0, 10) : "",
       practice_area: (record.practice_area || "") as PracticeArea,
-      appeal_subtype: record.appeal_subtype || "",
+      appeal_subtype: appealSubtypeLabel(record.appeal_subtype),
       source_type: (record.source_type || "") as SourceType,
       precedent_level: record.precedent_level || "",
       is_binding: record.is_binding ?? true,
@@ -84,7 +86,7 @@ export function PrecedentEditSheet({ caseLawId, onOpenChange }: Props) {
       headnote: record.headnote || "",
       key_quote: (record as { key_quote?: string }).key_quote || "",
     });
-  }, [record]);
+  }
 
   const onSubmit = async (e: React.FormEvent) => {
     e.preventDefault();

web-ui/src/lib/api/precedent-library.ts

@@ -46,6 +46,8 @@ export type Precedent = {
   headnote: string;
   subject_tags: string[];
   source_kind: string;
+  chair_name: string | null;
+  district: string | null;
   extraction_status: string;
   halacha_extraction_status: string;
   metadata_extraction_requested_at: string | null;
@@ -82,9 +84,20 @@ export type Halacha = {
   precedent_level?: string;
 };
 
+export type RelatedCase = {
+  id: string;
+  case_number: string;
+  case_name: string;
+  court: string;
+  precedent_level: string;
+  date: string | null;
+  relation_type: string;
+};
+
 export type PrecedentDetail = Precedent & {
   full_text: string;
   halachot: Halacha[];
+  related_cases: RelatedCase[];
 };
 
 export type SearchHit =
@@ -137,6 +150,7 @@ export type ListFilters = {
   court?: string;
   precedentLevel?: string;
   sourceType?: SourceType;
+  sourceKind?: string;
   search?: string;
   limit?: number;
   offset?: number;
@@ -164,6 +178,7 @@ export function usePrecedents(filters: ListFilters = {}) {
       if (filters.court) p.set("court", filters.court);
       if (filters.precedentLevel) p.set("precedent_level", filters.precedentLevel);
       if (filters.sourceType) p.set("source_type", filters.sourceType);
+      if (filters.sourceKind) p.set("source_kind", filters.sourceKind);
       if (filters.search) p.set("search", filters.search);
       if (filters.limit) p.set("limit", String(filters.limit));
       if (filters.offset) p.set("offset", String(filters.offset));
@@ -353,6 +368,40 @@ export function useDeletePrecedent() {
   });
 }
 
+export function useLinkRelatedCase(caseId: string) {
+  const qc = useQueryClient();
+  return useMutation({
+    mutationFn: (vars: { relatedId: string; relationType?: string }) =>
+      apiRequest<{ linked: boolean }>(
+        `/api/precedent-library/${encodeURIComponent(caseId)}/relations`,
+        {
+          method: "POST",
+          body: {
+            related_id: vars.relatedId,
+            relation_type: vars.relationType ?? "same_case_chain",
+          },
+        },
+      ),
+    onSuccess: () => {
+      qc.invalidateQueries({ queryKey: libraryKeys.detail(caseId) });
+    },
+  });
+}
+
+export function useUnlinkRelatedCase(caseId: string) {
+  const qc = useQueryClient();
+  return useMutation({
+    mutationFn: (relatedId: string) =>
+      apiRequest<{ unlinked: boolean }>(
+        `/api/precedent-library/${encodeURIComponent(caseId)}/relations/${encodeURIComponent(relatedId)}`,
+        { method: "DELETE" },
+      ),
+    onSuccess: () => {
+      qc.invalidateQueries({ queryKey: libraryKeys.detail(caseId) });
+    },
+  });
+}
+
 export type PrecedentPatch = Partial<{
   case_name: string;
   court: string;

web-ui/src/lib/api/types.ts

@@ -1355,7 +1355,7 @@ export interface paths {
         };
         /**
          * Api Case Agents
-         * @description Get all Paperclip agent activity for a case: issues, comments, agent status.
+         * @description Get all Paperclip agent activity for a case: issues, comments, interactions, agent status.
          */
         get: operations["api_case_agents_api_cases__case_number__agents_get"];
         put?: never;
@@ -1388,6 +1388,149 @@ export interface paths {
         patch?: never;
         trace?: never;
     };
+    "/api/cases/{case_number}/agents/interaction-response": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Api Post Interaction Response
+         * @description Submit a user's answer to a Paperclip issue-thread interaction.
+         *
+         *     Routes to /respond | /accept | /reject based on `action`. Paperclip
+         *     auto-wakes the issue assignee after a successful submission.
+         */
+        post: operations["api_post_interaction_response_api_cases__case_number__agents_interaction_response_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/api/settings/mcp/env": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Api Mcp Env
+         * @description List all catalog env vars with Coolify (authoritative) + container values.
+         */
+        get: operations["api_mcp_env_api_settings_mcp_env_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/api/settings/mcp/env/{key}": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        /**
+         * Api Mcp Env Update
+         * @description Update a non-secret env var in Coolify. Requires redeploy to take effect.
+         */
+        patch: operations["api_mcp_env_update_api_settings_mcp_env__key__patch"];
+        trace?: never;
+    };
+    "/api/settings/mcp/env/redeploy": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Api Mcp Env Redeploy
+         * @description Trigger Coolify redeploy of the legal-ai app.
+         */
+        post: operations["api_mcp_env_redeploy_api_settings_mcp_env_redeploy_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/api/settings/mcp/tools": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Api Mcp Tools
+         * @description List all MCP tools registered in legal_mcp.
+         */
+        get: operations["api_mcp_tools_api_settings_mcp_tools_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/api/settings/mcp/registrations": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Api Mcp Registrations
+         * @description List MCP server registrations from host config files.
+         */
+        get: operations["api_mcp_registrations_api_settings_mcp_registrations_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/api/settings/mcp/blocks": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Api Mcp Blocks
+         * @description List the 12-block decision schema (read-only reference).
+         */
+        get: operations["api_mcp_blocks_api_settings_mcp_blocks_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
     "/api/settings/paperclip-companies": {
         parameters: {
             query?: never;
@@ -1516,6 +1659,29 @@ export interface paths {
         patch?: never;
         trace?: never;
     };
+    "/api/admin/paperclip-agents": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Api List Paperclip Agents
+         * @description List all Paperclip agents grouped into master+mirror pairs with drift detection.
+         *
+         *     Read-only. Source of truth: Paperclip ``GET /api/companies/{id}/agents`` API
+         *     (not direct DB) — keeps us decoupled from Paperclip's schema changes.
+         */
+        get: operations["api_list_paperclip_agents_api_admin_paperclip_agents_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
     "/api/admin/skills/install": {
         parameters: {
             query?: never;
@@ -2295,6 +2461,27 @@ export interface components {
             /** Practice Areas */
             practice_areas?: string[] | null;
         };
+        /** InteractionResponseRequest */
+        InteractionResponseRequest: {
+            /** Issue Id */
+            issue_id: string;
+            /** Interaction Id */
+            interaction_id: string;
+            /**
+             * Action
+             * @enum {string}
+             */
+            action: "respond" | "accept" | "reject";
+            /** Payload */
+            payload: {
+                [key: string]: unknown;
+            };
+        };
+        /** McpEnvUpdateRequest */
+        McpEnvUpdateRequest: {
+            /** Value */
+            value: unknown;
+        };
         /** MethodologyUpdateRequest */
         MethodologyUpdateRequest: {
             /** Value */
@@ -4491,6 +4678,176 @@ export interface operations {
             };
         };
     };
+    api_post_interaction_response_api_cases__case_number__agents_interaction_response_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                case_number: string;
+            };
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["InteractionResponseRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    api_mcp_env_api_settings_mcp_env_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+        };
+    };
+    api_mcp_env_update_api_settings_mcp_env__key__patch: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                key: string;
+            };
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["McpEnvUpdateRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    api_mcp_env_redeploy_api_settings_mcp_env_redeploy_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+        };
+    };
+    api_mcp_tools_api_settings_mcp_tools_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+        };
+    };
+    api_mcp_registrations_api_settings_mcp_registrations_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+        };
+    };
+    api_mcp_blocks_api_settings_mcp_blocks_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+        };
+    };
     api_paperclip_companies_api_settings_paperclip_companies_get: {
         parameters: {
             query?: never;
@@ -4714,6 +5071,26 @@ export interface operations {
             };
         };
     };
+    api_list_paperclip_agents_api_admin_paperclip_agents_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+        };
+    };
     api_install_skill_api_admin_skills_install_post: {
         parameters: {
             query?: never;

web/app.py

@@ -20,7 +20,7 @@ sys.path.insert(0, str(Path(__file__).resolve().parent.parent / "mcp-server" / "
 
 import zipfile
 
-from fastapi import FastAPI, File, Form, HTTPException, UploadFile
+from fastapi import BackgroundTasks, FastAPI, File, Form, HTTPException, UploadFile
 from fastapi.responses import FileResponse, StreamingResponse
 from typing import Any, Literal
 from pydantic import BaseModel
@@ -44,7 +44,7 @@ from web.mcp_env_catalog import (
     normalize_for_compare,
 )
 from web.progress_store import ProgressStore
-from web.paperclip_api import pc_request
+from web.paperclip_api import emit_case_status_webhook, pc_request
 from web.paperclip_client import (
     COMPANIES as PAPERCLIP_COMPANIES,
     accept_interaction as pc_accept_interaction,
@@ -62,6 +62,7 @@ from web.paperclip_client import (
     respond_to_interaction as pc_respond_to_interaction,
     restore_project as pc_restore_project,
     wake_ceo_agent as pc_wake_ceo,
+    wake_curator_for_final as pc_wake_curator_for_final,
     wake_for_precedent_extraction as pc_wake_for_precedent_extraction,
 )
 
@@ -1336,8 +1337,12 @@ async def api_case_get(case_number: str):
 
 
 @app.put("/api/cases/{case_number}")
-async def api_case_update(case_number: str, req: CaseUpdateRequest):
+async def api_case_update(case_number: str, req: CaseUpdateRequest, background_tasks: BackgroundTasks):
     """Update case details."""
+    # Capture old status before the update so we can detect changes.
+    existing = await db.get_case_by_number(case_number)
+    old_status = (existing or {}).get("status", "")
+
     result = await cases_tools.case_update(
         case_number=case_number,
         status=req.status,
@@ -1350,10 +1355,30 @@ async def api_case_update(case_number: str, req: CaseUpdateRequest):
         expected_outcome=req.expected_outcome,
     )
     try:
-        return json.loads(result)
+        parsed = json.loads(result)
     except json.JSONDecodeError:
         raise HTTPException(404, result)
 
+    # Emit webhook when status changes (fire-and-forget via BackgroundTasks).
+    new_status = req.status
+    if new_status and old_status != new_status:
+        prefix = case_number[:1]
+        company_id = (
+            PAPERCLIP_COMPANIES["licensing"] if prefix == "1"
+            else PAPERCLIP_COMPANIES["betterment"] if prefix in ("8", "9")
+            else None
+        )
+        background_tasks.add_task(
+            emit_case_status_webhook,
+            case_number=case_number,
+            old_status=old_status,
+            new_status=new_status,
+            company_id=company_id,  # None is safe — plugin handles unknown company gracefully
+        )
+        logger.debug("webhook scheduled: case %s %s → %s", case_number, old_status, new_status)
+
+    return parsed
+
 
 @app.delete("/api/cases")
 async def api_case_delete(case_number: str, remove_files: bool = False):
@@ -2347,10 +2372,38 @@ async def api_mark_final(case_number: str, filename: str):
     if case_dir.exists():
         commit_and_push(case_dir, f"גרסה סופית: {final_name}")
 
+    # ingest_final_version into the case_law corpus is NOT called here:
+    # it uses claude_session under the hood, which only works when invoked
+    # from the local MCP server (Claude CLI present), not from this
+    # FastAPI container. Run it manually via Claude Code / MCP when needed.
+    # See: ~/.claude/projects/-home-chaim-legal-ai/memory/feedback_claude_session_local_only.md
+    ingest_status: dict = {"status": "skipped", "reason": "container_no_claude_cli"}
+
+    # Best-effort: wake the Knowledge Curator (Hermes) to analyze the
+    # signed final and propose updates to skills/lessons. Non-fatal on
+    # failure so marking final never breaks for the user.
+    curator_status: dict = {"status": "skipped"}
+    try:
+        # Company by case-number prefix: 1xxx=CMP (licensing), 8/9xxx=CMPA (betterment)
+        prefix = case_number[:1]
+        company_id = (
+            PAPERCLIP_COMPANIES["licensing"] if prefix == "1"
+            else PAPERCLIP_COMPANIES["betterment"] if prefix in ("8", "9")
+            else ""
+        )
+        curator_status = await pc_wake_curator_for_final(
+            case_number, final_name, company_id=company_id
+        )
+    except Exception as e:
+        logger.warning("curator wakeup failed for %s: %s", case_number, e)
+        curator_status = {"status": "error", "error": str(e)}
+
     return {
         "final_filename": final_name,
         "training_copy": str(training_dest),
         "status": "final",
+        "ingest_final": ingest_status,
+        "curator": curator_status,
     }
 
 
@@ -3028,8 +3081,16 @@ async def api_get_methodology(category: str):
     items = {}
     for key, default_val in defaults.items():
         if key in overrides:
+            raw = overrides[key]["rule_value"]
+            # asyncpg returns JSONB as a raw JSON string when no codec is registered.
+            # Parse it back to a Python object so the frontend receives the correct type.
+            if isinstance(raw, str):
+                try:
+                    raw = json.loads(raw)
+                except (json.JSONDecodeError, TypeError):
+                    pass
             items[key] = {
-                "value": overrides[key]["rule_value"],
+                "value": raw,
                 "is_override": True,
                 "updated_at": overrides[key]["created_at"].isoformat() if overrides[key]["created_at"] else None,
             }
@@ -3066,10 +3127,14 @@ async def api_update_methodology(category: str, key: str, req: MethodologyUpdate
             raise HTTPException(422, "content_checklists value must be a non-empty string")
 
     pool = await db.get_pool()
+    # json.dumps → text, then PostgreSQL casts text→jsonb.
+    # Passing a Python list directly causes "expected str, got list" in asyncpg;
+    # passing a str with ::jsonb causes double-encoding (stored as JSONB string).
+    # ::text::jsonb bypasses asyncpg's codec and lets PostgreSQL parse the JSON.
     await pool.execute(
         "INSERT INTO appeal_type_rules (id, appeal_type, rule_category, rule_key, rule_value) "
-        "VALUES (gen_random_uuid(), '_global', $1, $2, $3::jsonb) "
+        "VALUES (gen_random_uuid(), '_global', $1, $2, $3::text::jsonb) "
-        "ON CONFLICT (appeal_type, rule_category, rule_key) DO UPDATE SET rule_value = $3::jsonb",
+        "ON CONFLICT (appeal_type, rule_category, rule_key) DO UPDATE SET rule_value = $3::text::jsonb",
         category, key, json.dumps(req.value, ensure_ascii=False),
     )
 
@@ -3191,6 +3256,8 @@ _AGENT_NAME_ORDER = {
 }
 
 # Fields that should match between master (CMP) and mirror (CMPA). Drift = bug.
+# `status` is intentionally excluded — it's runtime state (running/idle/paused),
+# not config, and changes constantly.
 _DRIFT_FIELDS = (
     "model",
     "effort",
@@ -3204,10 +3271,21 @@ _DRIFT_FIELDS = (
     "wakeOnDemand",
     "maxConcurrentRuns",
     "budget_monthly_cents",
-    "status",
 )
 
 
+def _portable_skills(skills: list[str]) -> list[str]:
+    """Return only the skills whose drift across companies is meaningful.
+
+    `local/*` skills carry per-install hashes (different IDs per company even
+    when the underlying skill is identical); `company/{cid}/*` skills are
+    scoped to a single company by construction. Both are expected to differ
+    between master and mirror — comparing them produces noise. Only
+    `paperclipai/*` (vendor-shipped) skills should match exactly.
+    """
+    return sorted(s for s in skills if s.startswith("paperclipai/"))
+
+
 def _shape_paperclip_agent(raw: dict, company_id: str, company_name: str) -> dict:
     """Flatten a Paperclip agent row into the shape the UI consumes."""
     ac = raw.get("adapterConfig") or {}
@@ -3252,6 +3330,9 @@ def _compute_drift(master: dict | None, mirror: dict | None) -> list[dict]:
     for field in _DRIFT_FIELDS:
         m_val = master.get(field)
         i_val = mirror.get(field)
+        if field == "desiredSkills":
+            m_val = _portable_skills(m_val or [])
+            i_val = _portable_skills(i_val or [])
         if m_val != i_val:
             drift.append({"field": field, "master": m_val, "mirror": i_val})
     return drift
@@ -4246,13 +4327,15 @@ async def precedent_library_list(
     precedent_level: str = "",
     source_type: str = "",
     search: str = "",
+    source_kind: str = "external_upload",
     limit: int = 100,
     offset: int = 0,
 ):
     rows = await db.list_external_case_law(
         practice_area=practice_area, court=court,
         precedent_level=precedent_level, source_type=source_type,
-        search=search, limit=limit, offset=offset,
+        search=search, source_kind=source_kind,
+        limit=limit, offset=offset,
     )
     return {"items": rows, "count": len(rows)}
 
@@ -4333,6 +4416,37 @@ async def precedent_library_delete(case_law_id: str):
     return {"deleted": True, "case_law_id": case_law_id}
 
 
+class PrecedentRelationRequest(BaseModel):
+    related_id: str
+    relation_type: str = "same_case_chain"
+
+
+@app.post("/api/precedent-library/{case_law_id}/relations")
+async def precedent_add_relation(case_law_id: str, req: PrecedentRelationRequest):
+    try:
+        a = UUID(case_law_id)
+        b = UUID(req.related_id)
+    except ValueError:
+        raise HTTPException(400, "case_law_id לא תקין")
+    if not await db.get_case_law(a):
+        raise HTTPException(404, "פסיקה לא נמצאה")
+    if not await db.get_case_law(b):
+        raise HTTPException(404, f"פסיקה קשורה {req.related_id} לא נמצאה")
+    await db.add_case_law_relation(a, b, req.relation_type)
+    return {"linked": True, "case_law_id": case_law_id, "related_id": req.related_id}
+
+
+@app.delete("/api/precedent-library/{case_law_id}/relations/{related_id}")
+async def precedent_remove_relation(case_law_id: str, related_id: str):
+    try:
+        a = UUID(case_law_id)
+        b = UUID(related_id)
+    except ValueError:
+        raise HTTPException(400, "case_law_id לא תקין")
+    await db.remove_case_law_relation(a, b)
+    return {"unlinked": True, "case_law_id": case_law_id, "related_id": related_id}
+
+
 # Halacha and metadata extraction are LLM-driven and rely on the local
 # `claude` CLI via mcp-server/services/claude_session.py — they CANNOT run
 # from this container (no CLI, no claude.ai session). The endpoints below
@@ -4342,18 +4456,41 @@ async def precedent_library_delete(case_law_id: str):
 # drain the queue.
 
 
+async def _wake_ceo_for_precedent(case_law_id: UUID, kind: str) -> dict:
+    """Trigger Paperclip CEO to drain the precedent extraction queue, mirroring
+    the upload flow (see ``precedent_library_upload`` → ``pc_wake_for_precedent_extraction``).
+
+    Best-effort — any failure is logged but doesn't surface to the user, who
+    can still invoke ``mcp__legal-ai__precedent_process_pending`` manually.
+    Returns a dict with the wakeup outcome for inclusion in the API response.
+    """
+    record = await db.get_case_law(case_law_id)
+    if not record:
+        return {"ok": False, "skipped": "record_missing"}
+    try:
+        return await pc_wake_for_precedent_extraction(
+            case_law_id=str(case_law_id),
+            citation=str(record.get("case_number") or ""),
+            practice_area=str(record.get("practice_area") or ""),
+        )
+    except Exception:
+        logger.exception("precedent-extraction wakeup failed (non-fatal, kind=%s)", kind)
+        return {"ok": False, "error": "wakeup_failed"}
+
+
 @app.post("/api/precedent-library/{case_law_id}/request-metadata")
 async def precedent_request_metadata(case_law_id: str):
-    """Stamp the case_law row as needing metadata extraction. The local
+    """Stamp the case_law row as needing metadata extraction AND wake the
-    MCP worker (`precedent_process_pending_metadata`) will pick it up."""
+    Paperclip CEO so extraction runs automatically — same flow as upload."""
     try:
         cid = UUID(case_law_id)
     except ValueError:
         raise HTTPException(400, "case_law_id לא תקין")
     ok = await db.request_metadata_extraction(cid)
     if not ok:
-        raise HTTPException(404, "פסיקה לא נמצאה (או לא מסוג external_upload)")
+        raise HTTPException(404, "פסיקה לא נמצאה")
-    return {"queued": True, "case_law_id": case_law_id, "kind": "metadata"}
+    wakeup = await _wake_ceo_for_precedent(cid, kind="metadata")
+    return {"queued": True, "case_law_id": case_law_id, "kind": "metadata", "wakeup": wakeup}
 
 
 @app.post("/api/precedent-library/{case_law_id}/request-halachot")
@@ -4365,8 +4502,9 @@ async def precedent_request_halachot(case_law_id: str):
         raise HTTPException(400, "case_law_id לא תקין")
     ok = await db.request_halacha_extraction(cid)
     if not ok:
-        raise HTTPException(404, "פסיקה לא נמצאה (או לא מסוג external_upload)")
+        raise HTTPException(404, "פסיקה לא נמצאה")
-    return {"queued": True, "case_law_id": case_law_id, "kind": "halacha"}
+    wakeup = await _wake_ceo_for_precedent(cid, kind="halacha")
+    return {"queued": True, "case_law_id": case_law_id, "kind": "halacha", "wakeup": wakeup}
 
 
 @app.get("/api/precedent-library/queue/pending")
@@ -4379,6 +4517,148 @@ async def precedent_queue_pending(kind: str = "metadata", limit: int = 20):
     return {"items": items, "count": len(items)}
 
 
+from legal_mcp.services import internal_decisions as int_decisions_service  # noqa: E402
+
+
+@app.post("/api/internal-decisions/upload")
+async def internal_decisions_upload(
+    file: UploadFile = File(...),
+    case_number: str = Form(...),
+    case_name: str = Form(""),
+    court: str = Form(""),
+    decision_date: str = Form(""),
+    chair_name: str = Form(""),
+    district: str = Form(""),
+    practice_area: str = Form(""),
+    appeal_subtype: str = Form(""),
+    subject_tags: str = Form("[]"),
+    is_binding: bool = Form(True),
+    summary: str = Form(""),
+):
+    """Upload a planning appeals-committee decision to the internal corpus."""
+    if practice_area and practice_area not in _PRACTICE_AREAS:
+        raise HTTPException(400, "practice_area לא תקין")
+    if not case_number.strip():
+        raise HTTPException(400, "case_number חובה")
+
+    suffix = Path(file.filename or "").suffix.lower()
+    if suffix not in ALLOWED_EXTENSIONS:
+        raise HTTPException(400, f"סוג קובץ לא נתמך: {suffix}")
+
+    UPLOAD_DIR.mkdir(parents=True, exist_ok=True)
+    staged = UPLOAD_DIR / f"int_{uuid4().hex[:8]}_{file.filename}"
+    size = 0
+    with staged.open("wb") as out:
+        while chunk := await file.read(1024 * 1024):
+            size += len(chunk)
+            if size > MAX_FILE_SIZE:
+                staged.unlink(missing_ok=True)
+                raise HTTPException(413, "קובץ גדול מדי")
+            out.write(chunk)
+
+    try:
+        tags = json.loads(subject_tags) if subject_tags else []
+        if not isinstance(tags, list):
+            tags = []
+    except json.JSONDecodeError:
+        tags = []
+
+    task_id = str(uuid4())
+    await _progress.set(task_id, {
+        "status": "queued", "filename": file.filename or "",
+        "stage": "queued", "percent": 0,
+    })
+
+    async def _run():
+        try:
+            await int_decisions_service.ingest_internal_decision(
+                case_number=case_number.strip(),
+                case_name=case_name.strip(),
+                court=court.strip(),
+                decision_date=decision_date or None,
+                chair_name=chair_name.strip(),
+                district=district.strip(),
+                practice_area=practice_area,
+                appeal_subtype=appeal_subtype.strip(),
+                subject_tags=tags,
+                is_binding=is_binding,
+                summary=summary.strip(),
+                file_path=staged,
+            )
+            await _progress.set(task_id, {"status": "completed", "percent": 100})
+        except Exception as e:
+            logger.exception("internal-decisions upload failed")
+            await _progress.set(task_id, {"status": "failed", "error": str(e)})
+        finally:
+            staged.unlink(missing_ok=True)
+
+    asyncio.create_task(_run())
+    return {"task_id": task_id}
+
+
+@app.post("/api/internal-decisions/migrate")
+async def internal_decisions_migrate(
+    source: str = "both",
+    dry_run: bool = True,
+):
+    """Migrate existing data to the internal committee corpus.
+
+    source: 'style_corpus' | 'external_corpus' | 'both'
+    dry_run: if true, only report what would be done (no writes)
+    """
+    if source not in {"style_corpus", "external_corpus", "both"}:
+        raise HTTPException(400, "source חייב להיות style_corpus / external_corpus / both")
+
+    results: dict = {}
+    if source in {"style_corpus", "both"}:
+        results["style_corpus"] = await int_decisions_service.migrate_from_style_corpus(dry_run=dry_run)
+    if source in {"external_corpus", "both"}:
+        results["external_corpus"] = await int_decisions_service.migrate_from_external_corpus(dry_run=dry_run)
+    return results
+
+
+@app.get("/api/internal-decisions")
+async def internal_decisions_list(
+    district: str = "",
+    chair_name: str = "",
+    practice_area: str = "",
+    limit: int = 100,
+):
+    """List internal committee decisions with optional filters."""
+    pool = await db.get_pool()
+    async with pool.acquire() as conn:
+        conditions = ["source_kind = 'internal_committee'"]
+        params: list = []
+        idx = 1
+        if district:
+            conditions.append(f"district = ${idx}")
+            params.append(district)
+            idx += 1
+        if chair_name:
+            conditions.append(f"chair_name = ${idx}")
+            params.append(chair_name)
+            idx += 1
+        if practice_area:
+            conditions.append(f"practice_area = ${idx}")
+            params.append(practice_area)
+            idx += 1
+        params.append(limit)
+        where = " AND ".join(conditions)
+        rows = await conn.fetch(
+            f"SELECT id, case_number, case_name, court, district, chair_name, "
+            f"date, practice_area, appeal_subtype, extraction_status, halacha_extraction_status "
+            f"FROM case_law WHERE {where} ORDER BY date DESC NULLS LAST LIMIT ${idx}",
+            *params,
+        )
+        total = await conn.fetchval(
+            "SELECT COUNT(*) FROM case_law WHERE source_kind = 'internal_committee'"
+        )
+    return {
+        "total": total,
+        "items": [dict(r) for r in rows],
+    }
+
+
 @app.get("/api/halachot")
 async def halachot_list(
     case_law_id: str = "",

web/paperclip_api.py

@@ -19,6 +19,7 @@ from __future__ import annotations
 
 import logging
 import os
+from datetime import datetime
 from typing import Any
 
 import httpx
@@ -81,3 +82,35 @@ async def pc_request(
     if raise_on_error:
         resp.raise_for_status()
     return resp
+
+
+async def emit_case_status_webhook(
+    case_number: str,
+    old_status: str,
+    new_status: str,
+    company_id: str | None = None,
+    run_id: str | None = None,
+) -> None:
+    """Notify the Paperclip plugin that a case status changed.
+
+    Fire-and-forget: logs errors but never raises, so callers aren't blocked.
+    """
+    try:
+        await pc_request(
+            "POST",
+            "/api/plugins/marcusgroup.legal-ai/webhooks/case-status",
+            json={
+                "caseNumber": case_number,
+                "oldStatus": old_status,
+                "newStatus": new_status,
+                "companyId": company_id,
+                "timestamp": datetime.utcnow().isoformat() + "Z",
+            },
+            run_id=run_id,
+            timeout=5.0,
+        )
+    except Exception as exc:
+        logger.warning(
+            "emit_case_status_webhook failed for case %s (%s → %s): %s",
+            case_number, old_status, new_status, exc,
+        )

web/paperclip_client.py

@@ -45,6 +45,14 @@ CEO_AGENTS = {
 # Default for backwards compat
 CEO_AGENT_ID = CEO_AGENTS[COMPANIES["licensing"]]
 
+# Knowledge Curator (Hermes) agent per company — woken after a case is
+# marked final, to analyze the signed decision and propose updates to
+# the style guide / lessons. POC stage 1.
+CURATOR_AGENTS = {
+    COMPANIES["licensing"]: "60dce831-5c5b-4bae-bda9-5282d506f0dc",   # CMP curator
+    COMPANIES["betterment"]: "d6f7c55d-570a-46b8-8d72-1286d07da0d8",  # CMPA curator
+}
+
 # Fallback mapping — used only when DB lookup returns no results
 _FALLBACK_APPEAL_TYPE_TO_COMPANY = {
     "רישוי": COMPANIES["licensing"],
@@ -898,3 +906,92 @@ async def wake_ceo_agent(issue_id: str, case_number: str, company_id: str = "")
     result = resp.json()
     logger.info("CEO agent wakeup for case %s: %s", case_number, result)
     return result
+
+
+async def wake_curator_for_final(
+    case_number: str,
+    final_filename: str,
+    company_id: str = "",
+) -> dict:
+    """Wake the Knowledge Curator (Hermes) when a case is marked final.
+
+    Creates a child issue under the main case issue, assigns it to the
+    curator, and triggers wakeup. Best-effort — silently skips if no
+    curator is configured for the company or no main issue is found.
+
+    Returns ``{"status": "ok"|"skipped", ...}``.
+    """
+    if not PAPERCLIP_BOARD_API_KEY:
+        logger.warning("PAPERCLIP_BOARD_API_KEY not set — skipping curator wakeup")
+        return {"status": "skipped", "reason": "no_api_key"}
+
+    curator_id = CURATOR_AGENTS.get(company_id)
+    if not curator_id:
+        logger.info("No curator configured for company %s — skipping", company_id)
+        return {"status": "skipped", "reason": "no_curator", "company_id": company_id}
+
+    issues = await get_case_issues(case_number)
+    if not issues:
+        logger.warning("No Paperclip issues found for case %s — skipping curator", case_number)
+        return {"status": "skipped", "reason": "no_issue"}
+
+    main_issue = next((i for i in issues if i.get("status") == "in_progress"), None) or issues[0]
+    main_issue_id = main_issue["id"]
+
+    description = (
+        f"דפנה סימנה את ההחלטה הסופית של תיק {case_number} כסופית.\n"
+        f"קובץ סופי: `{final_filename}`\n\n"
+        f"סקור את ההחלטה מול skills/decision/SKILL.md ו-docs/legal-decision-lessons.md.\n"
+        f"חפש 3-5 דפוסי סגנון/דיון שלא תועדו. כתוב comment בעברית, ניטרלי, "
+        f"ממוספר. עדכן את MEMORY.md שלך. סגור את ה-issue (status=done)."
+    )
+    child_resp = await pc_request(
+        "POST",
+        f"/api/issues/{main_issue_id}/children",
+        json={
+            "title": f"[ערר {case_number}] סקירת ידע — Knowledge Curator",
+            "description": description,
+            "status": "in_progress",
+            "priority": "low",
+            "assigneeAgentId": curator_id,
+        },
+        raise_on_error=True,
+    )
+    sub_issue = child_resp.json()
+    sub_issue_id = sub_issue["id"]
+
+    # Tag plugin_state for case-number visibility on the case page
+    try:
+        conn = await asyncpg.connect(PAPERCLIP_DB_URL)
+        try:
+            await _link_case_to_issue(conn, sub_issue_id, case_number)
+        finally:
+            await conn.close()
+    except Exception as e:
+        logger.warning("plugin_state link failed for sub_issue=%s: %s", sub_issue_id, e)
+
+    # Trigger wakeup (use API per Paperclip rule — never DB insert)
+    wake_resp = await pc_request(
+        "POST",
+        f"/api/agents/{curator_id}/wakeup",
+        json={
+            "source": "on_demand",
+            "triggerDetail": "manual",
+            "reason": f"final_marked_{case_number}",
+            "payload": {
+                "issueId": sub_issue_id,
+                "mutation": "assignment",
+            },
+        },
+        raise_on_error=True,
+    )
+    logger.info(
+        "Curator wakeup for case %s: sub_issue=%s curator=%s wake=%s",
+        case_number, sub_issue_id, curator_id, wake_resp.status_code,
+    )
+    return {
+        "status": "ok",
+        "sub_issue_id": sub_issue_id,
+        "curator_id": curator_id,
+        "main_issue_id": main_issue_id,
+    }