Flatten cases directory structure and unify paths

- Remove cases/new|in-progress|completed subdivision (status managed in DB)
- Rename documents/original → documents/originals (consistent plural)
- Move exports from global data/exports/ into cases/{num}/exports/
- Add documents/research/ for case law and analysis files
- Update all agents, scripts, config, web API endpoints, and DB paths

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

mcp-server/src/legal_mcp/config.py

@@ -53,28 +53,15 @@ GOOGLE_CLOUD_VISION_API_KEY = os.environ.get("GOOGLE_CLOUD_VISION_API_KEY", "")
 # Data directory
 DATA_DIR = Path(os.environ.get("DATA_DIR", str(Path.home() / "legal-ai" / "data")))
 TRAINING_DIR = DATA_DIR / "training"
-EXPORTS_DIR = DATA_DIR / "exports"
+EXPORTS_DIR = DATA_DIR / "exports"  # legacy exports only
 
-# Cases directory — new structure: cases/{new,in-progress,completed}/{case_number}/
-CASES_BASE = Path(os.environ.get("CASES_BASE", str(Path.home() / "legal-ai" / "cases")))
-CASES_NEW = CASES_BASE / "new"
-CASES_IN_PROGRESS = CASES_BASE / "in-progress"
-CASES_COMPLETED = CASES_BASE / "completed"
-CASES_DIR = CASES_NEW  # backwards compatibility — new cases default here
-
-_STATUS_DIRS = [CASES_NEW, CASES_IN_PROGRESS, CASES_COMPLETED]
+# Cases directory — flat structure: data/cases/{case_number}/
+CASES_DIR = DATA_DIR / "cases"
 
 
 def find_case_dir(case_number: str) -> Path:
-    """Find a case directory across all status folders.
-
-    Returns the existing directory, or defaults to CASES_NEW/{case_number}.
-    """
-    for base in _STATUS_DIRS:
-        candidate = base / case_number
-        if candidate.exists():
-            return candidate
-    return CASES_NEW / case_number
+    """Return the case directory for a given case number."""
+    return CASES_DIR / case_number
 
 # Chunking parameters
 CHUNK_SIZE_TOKENS = 600

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

@@ -169,9 +169,9 @@ async def export_decision(case_id: UUID, output_path: str | None = None) -> str:
 
         _write_block_to_docx(doc, block_id, block["title"], content)
 
-    # Determine output path — versioned under data/exports/{case_number}/
+    # Determine output path — versioned under cases/{case_number}/exports/
     if not output_path:
-        export_dir = config.EXPORTS_DIR / case["case_number"]
+        export_dir = config.find_case_dir(case["case_number"]) / "exports"
         export_dir.mkdir(parents=True, exist_ok=True)
         # Find next version number
         existing = sorted(export_dir.glob("טיוטה-v*.docx"))

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

@@ -1,6 +1,6 @@
 """Lessons learned from comparing AI drafts to Dafna Tamir's final decisions.
 
-Source: /data/uploads/לקחים-לעדכון-שרת-כתיבת-החלטות.md
+Source: docs/legal-decision-lessons.md
 Based on analysis of: Hecht 1180-1181 (rejection) and Beit HaKerem 1126/25+1141/25 (partial acceptance).
 """
 

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

@@ -39,7 +39,7 @@ async def document_upload(
         title = source.stem
 
     # Copy file to case directory
-    case_dir = config.find_case_dir(case_number) / "documents"
+    case_dir = config.find_case_dir(case_number) / "documents" / "originals"
     case_dir.mkdir(parents=True, exist_ok=True)
     dest = case_dir / source.name
     shutil.copy2(str(source), str(dest))