Faulkner DB

# Multi-Project Markdown Scanner Configuration # Faulkner DB Knowledge Graph Ingestion System # # Path Configuration: # - Use "auto" for automatic detection based on script location # - Use "~" prefix for user home directory (e.g., ~/.claude/plans) # - Use absolute paths for custom locations # Project root directories to scan # Set to "auto" to scan the parent directory of faulkner-db project_paths: auto # Claude Code plan files directory claude_plans: enabled: true # Default: ~/.claude/plans (auto-detected) path: auto # Extract all sections from Claude plans extract_all_sections: true # Consolidate/deduplicate similar plans consolidate_duplicates: true # Patterns to exclude from scanning excluded_patterns: - venv - __pycache__ - .git - node_modules - ".*" # Hidden directories/files - dist - build - target - .pytest_cache - .mypy_cache # File age filter (only scan files modified in last N days) # Set to null to scan all files regardless of age file_age_filter_days: null # Maximum parallel workers for file processing max_workers: 6 # Dry run mode (show what would be scanned without processing) dry_run: false # Database path for scan tracking # Set to "auto" for automatic detection (PROJECT_ROOT/data/scanner_tracking.db) database_path: auto # Deduplication settings deduplication: enabled: true mode: smart # Options: none, exact, smart # Smart mode merges similar content but tracks all source files similarity_threshold: 0.85 # For fuzzy matching (0.0-1.0) # Progress reporting settings progress: enabled: true verbose: true # Show detailed file-by-file progress report_interval: 10 # Update progress every N files # Content extraction settings extraction: min_content_length: 50 # Minimum characters for extracted sections max_alternatives: 5 # Maximum alternative options to extract # Scan behavior scan_mode: incremental: true # Only scan new/modified files resume_on_failure: true # Resume interrupted scans skip_failed_files: true # Continue if individual files fail # Neo4j/FalkorDB connection (inherited from core config) # Used for storing extracted knowledge nodes graph_database: enabled: true host: localhost port: 6379 graph_name: knowledge_graph