signalwire-agents 0.1.47__tar.gz → 0.1.48__tar.gz

{signalwire_agents-0.1.47/signalwire_agents.egg-info → signalwire_agents-0.1.48}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: signalwire_agents
-Version: 0.1.47
+Version: 0.1.48
 Summary: SignalWire AI Agents SDK
 Author-email: SignalWire Team <info@signalwire.com>
 License: MIT

{signalwire_agents-0.1.47 → signalwire_agents-0.1.48}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "signalwire_agents"
-version = "0.1.47"
+version = "0.1.48"
 description = "SignalWire AI Agents SDK"
 authors = [
     {name = "SignalWire Team", email = "info@signalwire.com"}

{signalwire_agents-0.1.47 → signalwire_agents-0.1.48}/signalwire_agents/__init__.py

@@ -18,7 +18,7 @@ A package for building AI agents using SignalWire's AI and SWML capabilities.
 from .core.logging_config import configure_logging
 configure_logging()
 
-__version__ = "0.1.47"
+__version__ = "0.1.48"
 
 # Import core classes for easier access
 from .core.agent_base import AgentBase

{signalwire_agents-0.1.47 → signalwire_agents-0.1.48}/signalwire_agents/cli/build_search.py

@@ -10,6 +10,9 @@ See LICENSE file in the project root for full license information.
 import argparse
 import sys
 from pathlib import Path
+from datetime import datetime
+
+from signalwire_agents.search.models import MODEL_ALIASES, DEFAULT_MODEL, resolve_model_alias
 
 def main():
     """Main entry point for the build-search command"""
@@ -66,11 +69,35 @@ Examples:
   sw-search ./docs \\
     --chunking-strategy qa
 
+  # Model selection examples (performance vs quality tradeoff)
+  sw-search ./docs --model mini     # Fastest (~5x faster), 384 dims, good for most use cases
+  sw-search ./docs --model base     # Balanced speed/quality, 768 dims (previous default)
+  sw-search ./docs --model large    # Best quality (same as base currently)
+  # Or use full model names:
+  sw-search ./docs --model sentence-transformers/all-MiniLM-L6-v2
+  sw-search ./docs --model sentence-transformers/all-mpnet-base-v2
 
   # JSON-based chunking (pre-chunked content)
   sw-search ./api_chunks.json \
     --chunking-strategy json \
     --file-types json
+
+  # Export chunks to JSON for review (single file)
+  sw-search ./docs \\
+    --output-format json \\
+    --output all_chunks.json
+
+  # Export chunks to JSON (one file per source)
+  sw-search ./docs \\
+    --output-format json \\
+    --output-dir ./chunks/
+
+  # Build index from exported JSON chunks
+  sw-search ./chunks/ \\
+    --chunking-strategy json \\
+    --file-types json \\
+    --output final.swsearch
+
   # Full configuration example
   sw-search ./docs ./examples README.md \\
     --output ./knowledge.swsearch \\
@@ -95,6 +122,12 @@ Examples:
   sw-search remote http://localhost:8001 "how to create an agent" --index-name docs
   sw-search remote localhost:8001 "API reference" --index-name docs --count 3 --verbose
 
+  # Migrate between backends
+  sw-search migrate ./docs.swsearch --to-pgvector \\
+    --connection-string "postgresql://user:pass@localhost/db" \\
+    --collection-name docs_collection
+  sw-search migrate --info ./docs.swsearch
+
   # PostgreSQL pgvector backend
   sw-search ./docs \\
     --backend pgvector \\
@@ -126,6 +159,18 @@ Examples:
         help='Output .swsearch file (default: sources.swsearch) or collection name for pgvector'
     )
     
+    parser.add_argument(
+        '--output-dir',
+        help='Output directory for results (creates one file per source file when used with --output-format json, or auto-names index files)'
+    )
+    
+    parser.add_argument(
+        '--output-format',
+        choices=['index', 'json'],
+        default='index',
+        help='Output format: index (create search index) or json (export chunks as JSON) (default: index)'
+    )
+    
     parser.add_argument(
         '--backend',
         choices=['sqlite', 'pgvector'],
@@ -197,8 +242,8 @@ Examples:
     
     parser.add_argument(
         '--model', 
-        default='sentence-transformers/all-mpnet-base-v2',
-        help='Sentence transformer model name (default: sentence-transformers/all-mpnet-base-v2)'
+        default=DEFAULT_MODEL,
+        help=f'Sentence transformer model name or alias (mini/base/large). Default: mini ({DEFAULT_MODEL})'
     )
     
     parser.add_argument(
@@ -241,6 +286,9 @@ Examples:
     
     args = parser.parse_args()
     
+    # Resolve model aliases
+    args.model = resolve_model_alias(args.model)
+    
     # Validate sources
     valid_sources = []
     for source in args.sources:
@@ -259,8 +307,35 @@ Examples:
         print("Error: --connection-string is required for pgvector backend")
         sys.exit(1)
     
-    # Default output filename
-    if not args.output:
+    # Validate output options
+    if args.output and args.output_dir:
+        print("Error: Cannot specify both --output and --output-dir")
+        sys.exit(1)
+    
+    # Handle JSON output format differently
+    if args.output_format == 'json':
+        # JSON export doesn't use backend
+        if args.backend != 'sqlite':
+            print("Warning: --backend is ignored when using --output-format json")
+        
+        # Determine output location
+        if args.output_dir:
+            # Multiple files mode
+            output_path = Path(args.output_dir)
+            if not output_path.exists():
+                output_path.mkdir(parents=True, exist_ok=True)
+        elif args.output:
+            # Single file mode
+            output_path = Path(args.output)
+            if not output_path.suffix:
+                output_path = output_path.with_suffix('.json')
+        else:
+            # Default to single file
+            output_path = Path('chunks.json')
+            args.output = str(output_path)
+    
+    # Default output filename (for index format)
+    if args.output_format == 'index' and not args.output and not args.output_dir:
         if args.backend == 'sqlite':
             if len(valid_sources) == 1:
                 # Single source - use its name
@@ -277,8 +352,25 @@ Examples:
             else:
                 args.output = "documents"
     
-    # Ensure output has .swsearch extension for sqlite
-    if args.backend == 'sqlite' and not args.output.endswith('.swsearch'):
+    # Handle --output-dir for index format
+    if args.output_format == 'index' and args.output_dir:
+        # Auto-generate output filename in the directory
+        if len(valid_sources) == 1:
+            source_name = valid_sources[0].stem if valid_sources[0].is_file() else valid_sources[0].name
+        else:
+            source_name = "combined"
+        
+        output_dir = Path(args.output_dir)
+        output_dir.mkdir(parents=True, exist_ok=True)
+        
+        if args.backend == 'sqlite':
+            args.output = str(output_dir / f"{source_name}.swsearch")
+        else:
+            # For pgvector, still use the name as collection
+            args.output = source_name
+    
+    # Ensure output has .swsearch extension for sqlite (but not for JSON format)
+    if args.output_format == 'index' and args.backend == 'sqlite' and args.output and not args.output.endswith('.swsearch'):
         args.output += '.swsearch'
     
     # Parse lists
@@ -325,6 +417,103 @@ Examples:
         print()
     
     try:
+        # Handle JSON export mode
+        if args.output_format == 'json':
+            # Import what we need for chunking
+            from signalwire_agents.search.index_builder import IndexBuilder
+            import json
+            
+            builder = IndexBuilder(
+                chunking_strategy=args.chunking_strategy,
+                max_sentences_per_chunk=args.max_sentences_per_chunk,
+                chunk_size=args.chunk_size,
+                chunk_overlap=args.overlap_size,
+                split_newlines=args.split_newlines,
+                index_nlp_backend=args.index_nlp_backend,
+                verbose=args.verbose,
+                semantic_threshold=args.semantic_threshold,
+                topic_threshold=args.topic_threshold
+            )
+            
+            # Process files and export chunks
+            all_chunks = []
+            chunk_files_created = []
+            
+            # Discover files from sources
+            files = builder._discover_files_from_sources(valid_sources, file_types, exclude_patterns)
+            
+            if args.verbose:
+                print(f"Processing {len(files)} files...")
+            
+            for file_path in files:
+                try:
+                    # Determine base directory for relative paths
+                    base_dir = builder._get_base_directory_for_file(file_path, valid_sources)
+                    
+                    # Process file into chunks
+                    chunks = builder._process_file(file_path, base_dir, tags)
+                    
+                    if args.output_dir:
+                        # Create individual JSON file
+                        relative_path = file_path.relative_to(base_dir) if base_dir else file_path.name
+                        json_filename = relative_path.with_suffix('.json')
+                        json_path = Path(args.output_dir) / json_filename
+                        
+                        # Create subdirectories if needed
+                        json_path.parent.mkdir(parents=True, exist_ok=True)
+                        
+                        # Save chunks to JSON
+                        chunk_data = {
+                            "chunks": chunks,
+                            "metadata": {
+                                "source_file": str(relative_path),
+                                "total_chunks": len(chunks),
+                                "chunking_strategy": args.chunking_strategy,
+                                "processing_date": datetime.now().isoformat()
+                            }
+                        }
+                        
+                        with open(json_path, 'w', encoding='utf-8') as f:
+                            json.dump(chunk_data, f, indent=2, ensure_ascii=False)
+                        
+                        chunk_files_created.append(json_path)
+                        if args.verbose:
+                            print(f"  Created: {json_path} ({len(chunks)} chunks)")
+                    else:
+                        # Accumulate all chunks for single file output
+                        all_chunks.extend(chunks)
+                        
+                except Exception as e:
+                    print(f"Error processing {file_path}: {e}")
+                    if args.verbose:
+                        import traceback
+                        traceback.print_exc()
+            
+            # Handle single file output
+            if not args.output_dir:
+                output_data = {
+                    "chunks": all_chunks,
+                    "metadata": {
+                        "total_chunks": len(all_chunks),
+                        "total_files": len(files),
+                        "chunking_strategy": args.chunking_strategy,
+                        "processing_date": datetime.now().isoformat()
+                    }
+                }
+                
+                with open(args.output, 'w', encoding='utf-8') as f:
+                    json.dump(output_data, f, indent=2, ensure_ascii=False)
+                
+                print(f"✓ Exported {len(all_chunks)} chunks to {args.output}")
+            else:
+                print(f"✓ Created {len(chunk_files_created)} JSON files in {args.output_dir}")
+                total_chunks = sum(len(json.load(open(f))['chunks']) for f in chunk_files_created)
+                print(f"  Total chunks: {total_chunks}")
+            
+            # Exit early for JSON format
+            return
+        
+        # Regular index building mode
         # Create index builder - import only when actually needed
         from signalwire_agents.search.index_builder import IndexBuilder
         builder = IndexBuilder(
@@ -370,7 +559,13 @@ Examples:
                 sys.exit(1)
         
         if args.backend == 'sqlite':
-            print(f"\n✓ Search index created successfully: {args.output}")
+            # Check if the index was actually created
+            import os
+            if os.path.exists(args.output):
+                print(f"\n✓ Search index created successfully: {args.output}")
+            else:
+                print(f"\n✗ Search index creation failed - no files were processed")
+                sys.exit(1)
         else:
             print(f"\n✓ Search collection created successfully: {args.output}")
             print(f"   Connection: {args.connection_string}")
@@ -427,21 +622,41 @@ def search_command():
     """Search within an existing search index"""
     parser = argparse.ArgumentParser(description='Search within a .swsearch index file or pgvector collection')
     parser.add_argument('index_source', help='Path to .swsearch file or collection name for pgvector')
-    parser.add_argument('query', help='Search query')
+    parser.add_argument('query', nargs='?', help='Search query (optional if using --shell)')
     parser.add_argument('--backend', choices=['sqlite', 'pgvector'], default='sqlite',
                        help='Storage backend (default: sqlite)')
     parser.add_argument('--connection-string', help='PostgreSQL connection string for pgvector backend')
+    parser.add_argument('--shell', action='store_true', 
+                       help='Interactive shell mode - load once and search multiple times')
     parser.add_argument('--count', type=int, default=5, help='Number of results to return (default: 5)')
     parser.add_argument('--distance-threshold', type=float, default=0.0, help='Minimum similarity score (default: 0.0)')
     parser.add_argument('--tags', help='Comma-separated tags to filter by')
     parser.add_argument('--query-nlp-backend', choices=['nltk', 'spacy'], default='nltk', 
                        help='NLP backend for query processing: nltk (fast, default) or spacy (better quality, slower)')
+    parser.add_argument('--keyword-weight', type=float, default=None,
+                       help='Manual keyword weight (0.0-1.0). Overrides automatic weight detection.')
     parser.add_argument('--verbose', action='store_true', help='Show detailed information')
     parser.add_argument('--json', action='store_true', help='Output results as JSON')
     parser.add_argument('--no-content', action='store_true', help='Hide content in results (show only metadata)')
+    parser.add_argument('--model', help='Override embedding model for query (mini/base/large or full model name)')
     
     args = parser.parse_args()
     
+    # Validate arguments
+    if not args.shell and not args.query:
+        print("Error: Query is required unless using --shell mode")
+        sys.exit(1)
+    
+    # Resolve model aliases
+    if args.model and args.model in MODEL_ALIASES:
+        args.model = MODEL_ALIASES[args.model]
+    
+    # Validate keyword weight if provided
+    if args.keyword_weight is not None:
+        if args.keyword_weight < 0.0 or args.keyword_weight > 1.0:
+            print("Error: --keyword-weight must be between 0.0 and 1.0")
+            sys.exit(1)
+    
     # Validate backend configuration
     if args.backend == 'pgvector' and not args.connection_string:
         print("Error: --connection-string is required for pgvector backend")
@@ -469,21 +684,167 @@ def search_command():
                 print(f"Connecting to pgvector collection: {args.index_source}")
         
         if args.backend == 'sqlite':
-            engine = SearchEngine(backend='sqlite', index_path=args.index_source)
+            # Pass the model from the index or override if specified
+            model = args.model if args.model else None
+            engine = SearchEngine(backend='sqlite', index_path=args.index_source, model=model)
         else:
+            # Pass the model override if specified
+            model = args.model if args.model else None
             engine = SearchEngine(backend='pgvector', connection_string=args.connection_string,
-                                collection_name=args.index_source)
+                                collection_name=args.index_source, model=model)
         
         # Get index stats
         stats = engine.get_stats()
+        
+        # Get the model from index config if not overridden
+        model_to_use = args.model
+        if not model_to_use and 'config' in stats:
+            # SQLite uses 'embedding_model', pgvector uses 'model_name'
+            model_to_use = stats['config'].get('embedding_model') or stats['config'].get('model_name')
+            
+        # Shell mode implementation
+        if args.shell:
+            import time
+            print(f"Search Shell - Index: {args.index_source}")
+            print(f"Backend: {args.backend}")
+            print(f"Index contains {stats['total_chunks']} chunks from {stats['total_files']} files")
+            if model_to_use:
+                print(f"Model: {model_to_use}")
+            print("Type 'exit' or 'quit' to leave, 'help' for options")
+            print("-" * 60)
+            
+            while True:
+                try:
+                    query = input("\nsearch> ").strip()
+                    
+                    if not query:
+                        continue
+                        
+                    if query.lower() in ['exit', 'quit', 'q']:
+                        print("Goodbye!")
+                        break
+                        
+                    if query.lower() == 'help':
+                        print("\nShell commands:")
+                        print("  help           - Show this help")
+                        print("  exit/quit/q    - Exit shell")
+                        print("  count=N        - Set result count (current: {})".format(args.count))
+                        print("  tags=tag1,tag2 - Set tag filter (current: {})".format(args.tags or 'none'))
+                        print("  verbose        - Toggle verbose output")
+                        print("\nOr type any search query...")
+                        continue
+                    
+                    # Handle shell commands
+                    if query.startswith('count='):
+                        try:
+                            args.count = int(query.split('=')[1])
+                            print(f"Result count set to: {args.count}")
+                        except:
+                            print("Invalid count value")
+                        continue
+                        
+                    if query.startswith('tags='):
+                        tag_str = query.split('=', 1)[1]
+                        args.tags = tag_str if tag_str else None
+                        tags = [tag.strip() for tag in args.tags.split(',')] if args.tags else None
+                        print(f"Tags filter set to: {tags or 'none'}")
+                        continue
+                        
+                    if query == 'verbose':
+                        args.verbose = not args.verbose
+                        print(f"Verbose output: {'on' if args.verbose else 'off'}")
+                        continue
+                    
+                    # Perform search with timing
+                    start_time = time.time()
+                    
+                    # Preprocess query
+                    enhanced = preprocess_query(
+                        query,
+                        vector=True,
+                        query_nlp_backend=args.query_nlp_backend,
+                        model_name=model_to_use,
+                        preserve_original=True,
+                        max_synonyms=2
+                    )
+                    
+                    # Parse tags
+                    tags = [tag.strip() for tag in args.tags.split(',')] if args.tags else None
+                    
+                    # Perform search
+                    results = engine.search(
+                        query_vector=enhanced.get('vector'),
+                        enhanced_text=enhanced.get('enhanced_text', query),
+                        count=args.count,
+                        distance_threshold=args.distance_threshold,
+                        tags=tags,
+                        keyword_weight=args.keyword_weight,
+                        original_query=query
+                    )
+                    
+                    search_time = time.time() - start_time
+                    
+                    # Display results
+                    if not results:
+                        print(f"\nNo results found for '{query}' ({search_time:.3f}s)")
+                    else:
+                        print(f"\nFound {len(results)} result(s) for '{query}' ({search_time:.3f}s):")
+                        if enhanced.get('enhanced_text') != query and args.verbose:
+                            print(f"Enhanced query: '{enhanced.get('enhanced_text')}'")
+                        print("=" * 60)
+                        
+                        for i, result in enumerate(results):
+                            print(f"\n[{i+1}] Score: {result['score']:.4f}")
+                            
+                            # Show metadata
+                            metadata = result['metadata']
+                            print(f"File: {metadata.get('filename', 'Unknown')}")
+                            if metadata.get('section'):
+                                print(f"Section: {metadata['section']}")
+                            
+                            # Show content unless suppressed
+                            if not args.no_content:
+                                content = result['content']
+                                if len(content) > 300 and not args.verbose:
+                                    content = content[:300] + "..."
+                                print(f"\n{content}")
+                            
+                            if i < len(results) - 1:
+                                print("-" * 40)
+                
+                except KeyboardInterrupt:
+                    print("\nUse 'exit' to quit")
+                except EOFError:
+                    print("\nGoodbye!")
+                    break
+                except Exception as e:
+                    print(f"\nError: {e}")
+                    if args.verbose:
+                        import traceback
+                        traceback.print_exc()
+            
+            return  # Exit after shell mode
+        
+        # Normal single query mode
         if args.verbose:
             print(f"Index contains {stats['total_chunks']} chunks from {stats['total_files']} files")
             print(f"Searching for: '{args.query}'")
             print(f"Query NLP Backend: {args.query_nlp_backend}")
+            if args.model:
+                print(f"Override model: {args.model}")
+            elif model_to_use:
+                print(f"Using index model: {model_to_use}")
             print()
         
         # Preprocess query
-        enhanced = preprocess_query(args.query, vector=True, query_nlp_backend=args.query_nlp_backend)
+        enhanced = preprocess_query(
+            args.query, 
+            vector=True,  # Both backends need vector for similarity search
+            query_nlp_backend=args.query_nlp_backend, 
+            model_name=model_to_use,
+            preserve_original=True,  # Keep original query terms
+            max_synonyms=2  # Reduce synonym expansion
+        )
         
         # Parse tags if provided
         tags = [tag.strip() for tag in args.tags.split(',')] if args.tags else None
@@ -494,7 +855,9 @@ def search_command():
             enhanced_text=enhanced.get('enhanced_text', args.query),
             count=args.count,
             distance_threshold=args.distance_threshold,
-            tags=tags
+            tags=tags,
+            keyword_weight=args.keyword_weight,
+            original_query=args.query  # Pass original for exact match boosting
         )
         
         if args.json:
@@ -563,6 +926,142 @@ def search_command():
             traceback.print_exc()
         sys.exit(1)
 
+def migrate_command():
+    """Migrate search indexes between backends"""
+    parser = argparse.ArgumentParser(
+        description='Migrate search indexes between SQLite and pgvector backends',
+        epilog="""
+Examples:
+  # Migrate SQLite to pgvector
+  sw-search migrate ./docs.swsearch \\
+    --to-pgvector \\
+    --connection-string "postgresql://user:pass@localhost/db" \\
+    --collection-name docs_collection
+  
+  # Migrate with overwrite
+  sw-search migrate ./docs.swsearch \\
+    --to-pgvector \\
+    --connection-string "postgresql://user:pass@localhost/db" \\
+    --collection-name docs_collection \\
+    --overwrite
+  
+  # Get index information
+  sw-search migrate --info ./docs.swsearch
+        """,
+        formatter_class=argparse.RawDescriptionHelpFormatter
+    )
+    
+    # Source argument (optional if using --info)
+    parser.add_argument('source', nargs='?', help='Source index file or collection')
+    
+    # Migration direction
+    migration_group = parser.add_mutually_exclusive_group()
+    migration_group.add_argument('--to-pgvector', action='store_true',
+                                help='Migrate SQLite index to pgvector')
+    migration_group.add_argument('--to-sqlite', action='store_true',
+                                help='Migrate pgvector collection to SQLite (not yet implemented)')
+    migration_group.add_argument('--info', action='store_true',
+                                help='Show information about an index')
+    
+    # pgvector options
+    parser.add_argument('--connection-string', 
+                       help='PostgreSQL connection string for pgvector')
+    parser.add_argument('--collection-name',
+                       help='Collection name for pgvector')
+    parser.add_argument('--overwrite', action='store_true',
+                       help='Overwrite existing collection')
+    
+    # SQLite options
+    parser.add_argument('--output',
+                       help='Output .swsearch file path (for --to-sqlite)')
+    
+    # Common options
+    parser.add_argument('--batch-size', type=int, default=100,
+                       help='Number of chunks to process at once (default: 100)')
+    parser.add_argument('--verbose', action='store_true',
+                       help='Show detailed progress')
+    
+    args = parser.parse_args()
+    
+    # Handle --info flag
+    if args.info:
+        if not args.source:
+            print("Error: Source index required with --info")
+            sys.exit(1)
+            
+        try:
+            from signalwire_agents.search.migration import SearchIndexMigrator
+            migrator = SearchIndexMigrator(verbose=args.verbose)
+            info = migrator.get_index_info(args.source)
+            
+            print(f"Index Information: {args.source}")
+            print(f"  Type: {info['type']}")
+            if info['type'] == 'sqlite':
+                print(f"  Total chunks: {info['total_chunks']}")
+                print(f"  Total files: {info['total_files']}")
+                print(f"  Model: {info['config'].get('embedding_model', 'Unknown')}")
+                print(f"  Dimensions: {info['config'].get('embedding_dimensions', 'Unknown')}")
+                print(f"  Created: {info['config'].get('created_at', 'Unknown')}")
+                if args.verbose:
+                    print("\n  Full configuration:")
+                    for key, value in info['config'].items():
+                        print(f"    {key}: {value}")
+            else:
+                print("  Unable to determine index type")
+        except Exception as e:
+            print(f"Error getting index info: {e}")
+            sys.exit(1)
+        return
+    
+    # Validate arguments for migration
+    if not args.source:
+        print("Error: Source index required for migration")
+        sys.exit(1)
+    
+    if not args.to_pgvector and not args.to_sqlite:
+        print("Error: Must specify migration direction (--to-pgvector or --to-sqlite)")
+        sys.exit(1)
+    
+    try:
+        from signalwire_agents.search.migration import SearchIndexMigrator
+        migrator = SearchIndexMigrator(verbose=args.verbose)
+        
+        if args.to_pgvector:
+            # Validate pgvector arguments
+            if not args.connection_string:
+                print("Error: --connection-string required for pgvector migration")
+                sys.exit(1)
+            if not args.collection_name:
+                print("Error: --collection-name required for pgvector migration")
+                sys.exit(1)
+            
+            # Perform migration
+            print(f"Migrating {args.source} to pgvector collection '{args.collection_name}'...")
+            stats = migrator.migrate_sqlite_to_pgvector(
+                sqlite_path=args.source,
+                connection_string=args.connection_string,
+                collection_name=args.collection_name,
+                overwrite=args.overwrite,
+                batch_size=args.batch_size
+            )
+            
+            print(f"\n✓ Migration completed successfully!")
+            print(f"  Chunks migrated: {stats['chunks_migrated']}")
+            print(f"  Errors: {stats['errors']}")
+            
+        elif args.to_sqlite:
+            print("Error: pgvector to SQLite migration not yet implemented")
+            print("This feature is planned for future development")
+            sys.exit(1)
+            
+    except Exception as e:
+        print(f"\nError during migration: {e}")
+        if args.verbose:
+            import traceback
+            traceback.print_exc()
+        sys.exit(1)
+
+
 def remote_command():
     """Search via remote API endpoint"""
     parser = argparse.ArgumentParser(description='Search via remote API endpoint')
@@ -838,6 +1337,11 @@ Examples:
             sys.argv.pop(1)
             remote_command()
             return
+        elif sys.argv[1] == 'migrate':
+            # Remove 'migrate' from argv and call migrate_command
+            sys.argv.pop(1)
+            migrate_command()
+            return
     
     # Regular build command
     main()