Memory

Search first: /mnemonic:search {relevant_keywords} Capture after: /mnemonic:capture {namespace} "{title}"

Run /mnemonic:list --namespaces to see available namespaces from loaded ontologies.

Ontology Skill

Provides custom ontology support for extending mnemonic with domain-specific knowledge structures.

Requirements

• PyYAML - Optional but recommended for YAML parsing. The library works without it but with limited functionality.

Capabilities

• Load and validate ontology definitions (YAML)
• Define custom namespaces beyond the 9 base namespaces
• Create typed entities (semantic, episodic, procedural)
• Establish relationships between entities
• Use traits/mixins for reusable field sets
• Entity discovery patterns

Cognitive Memory Type Triad

All entity types inherit from one of three base memory types:

Usage

ontology_validator.py

Validate ontology YAML files against the schema.

python ${SKILL_DIR}/lib/ontology_validator.py <file> [--json]

ontology_registry.py

Load and query ontologies.

python ${SKILL_DIR}/lib/ontology_registry.py [OPTIONS]

entity_resolver.py

Resolve and search entity references across memories.

python ${SKILL_DIR}/lib/entity_resolver.py [OPTIONS]

Entity Reference Syntax

• Simple: @[[Entity Name]] - Resolves by name
• Typed: [[technology:postgres-id]] - Resolves by type and ID

Discovery Patterns

Ontologies can define patterns for automatic namespace suggestion during memory capture.

Content Patterns

Match text content to suggest appropriate namespaces:

discovery:
  content_patterns:
    - pattern: "\\bdecided to\\b|\\bwe will use\\b"
      namespace: _semantic/decisions
    - pattern: "\\blearned that\\b|\\bthe fix was\\b"
      namespace: _semantic/knowledge

File Patterns

Match file paths for contextual namespace hints:

discovery:
  file_patterns:
    - pattern: "auth|login|session"
      namespaces:
        - _semantic/knowledge
        - _semantic/decisions
      context: authentication

Configuration

discovery:
  enabled: true
  confidence_threshold: 0.8  # Minimum confidence to suggest

See fallback/ontologies/mif-base.ontology.yaml for complete examples.

Files

• lib/ontology_registry.py - Load and manage ontologies
• lib/ontology_validator.py - Validate ontology YAML
• lib/ontology_loader.py - Centralized ontology loading with caching
• lib/entity_resolver.py - Resolve entity references
• fallback/ontologies/mif-base.ontology.yaml - Standard mnemonic namespaces
• fallback/schema/ontology/ontology.schema.json - JSON Schema