Vectors Command

Manage vector databases for semantic search and RAG (Retrieval-Augmented Generation) applications. The vectors command provides database lifecycle management including creation, deletion, and inspection.

Overview

Vector databases store embeddings generated from text documents, enabling semantic search capabilities. The vectors command helps manage these databases, view their contents, and maintain storage efficiency.

Usage

# List all vector databases
lc vectors list

# Show database information
lc vectors info <database>

# Delete a database
lc vectors delete <database>

# Using aliases
lc v list
lc v info docs
lc v delete old-db

Subcommands

Name	Alias	Description
list	l	List all vector databases
delete	d	Delete a vector database
info	i	Show information about a database

Options

Short	Long	Description	Default
-h	--help	Print help	False

Examples

Database Management

List Databases

lc vectors list
# Output:
#   • project-docs (1,247 vectors)
#   • knowledge-base (523 vectors)
#   • research-papers (89 vectors)

# Short form
lc v l

Database Information

lc vectors info project-docs
# Output:
# Database: project-docs
# Vectors: 1,247
# Dimensions: 1536
# Model: text-embedding-3-small
# Size: 12.3 MB
# Created: 2024-01-15
# Last updated: 2024-01-20

lc v i project-docs

Delete Database

lc vectors delete old-project
# Will prompt for confirmation

lc v d old-project

Complete RAG Workflow

# Step 1: Create embeddings
lc embed -f "docs/*.md" --vectordb project-docs

# Step 2: Verify database
lc vectors info project-docs

# Step 3: Search similar content
lc similar -v project-docs "deployment process"

# Step 4: Use in chat
lc -v project-docs "How do I deploy the application?"

# Step 5: Cleanup when done
lc vectors delete project-docs

See Also

• Embed Command
• Similar Command

Troubleshooting

Common Issues

"Database not found"

• Error: Specified vector database doesn't exist
• Solution: Use lc vectors list to see available databases
• Create: Use lc embed -v <name> to create new database

"Permission denied"

• Error: Cannot access vector database files
• Solution: Check file permissions on database directory
• Fix: chmod 755 ~/.config/lc/vectors/

"Database corrupt"

• Error: Vector database file is corrupted
• Solution: Delete and recreate the database
• Backup: Export important data before deletion

Best Practices

1. Consistent Models: Use the same embedding model for all vectors in a database
2. Regular Cleanup: Remove unused databases to save disk space
3. Meaningful Names: Use descriptive names for databases
4. Size Management: Monitor database sizes for performance

Performance Considerations

• Large databases (>10k vectors) may have slower search times
• Consider splitting large databases by topic or date
• Regular maintenance improves query performance
• Monitor disk space usage

# Check database sizes
du -sh ~/.config/lc/vectors/*

# Performance monitoring
lc vectors list  # Shows vector counts
lc vectors info <db>  # Shows detailed stats

Database Location

Vector databases are stored locally:

• Linux/macOS: ~/.config/lc/vectors/
• Windows: %APPDATA%\lc\vectors\

Each database is a directory containing:

• Vector data files
• Metadata and configuration
• Search indices

Security and Backup

# Backup vector databases
tar -czf vectors-backup.tar.gz ~/.config/lc/vectors/

# Restore from backup
cd ~/.config/lc/
tar -xzf vectors-backup.tar.gz

# Secure database directory
chmod 700 ~/.config/lc/vectors/