memory-taxonomy

name	memory-taxonomy
description	Education domain memory taxonomy for concept entity resolution, predicates, permanence, tags, and example fact patterns.
version	1.0.0
tools_required	["memory_store_fact","mind_map_node_create","mind_map_node_get","curriculum_next_node","spaced_repetition_pending_reviews"]

Education Memory Taxonomy Skill

Purpose

Load this skill when storing education-domain memory facts or resolving a mind map concept entity for those facts.

Entity Resolution for Education Concepts

Every mind map node has an entity_id field backed by public.entities. This entity uniquely identifies the concept across the butler system and enables memory deduplication: facts stored with entity_id are linked to the canonical entity rather than relying on free-text subject matching.

Canonical name pattern: '<map_title> > <label>' For example, a node labelled "list comprehensions" in the "Python" map has canonical name "Python > list comprehensions".

Where to find entity_id:

• mind_map_node_create(): returned in the response dict as entity_id
• mind_map_node_get(): included in the node dict
• curriculum_next_node(): included in the node dict
• spaced_repetition_pending_reviews(): does NOT return entity_id; call mind_map_node_get(node_id=<id>) for each due node to retrieve its entity_id

Always pass entity_id to memory_store_fact() for concept-level facts (learning_outcome, struggle_area, prerequisite_mastered). This ensures facts are linked to the correct entity and not silently duplicated by subject-string variation.

Topic-level and user-level facts (study_pattern, learning_preference with subject="user") may use entity_id when a relevant map-level entity is available, but it is not required for those predicates.

Education Domain Taxonomy

Subject:

• For topic-level knowledge: topic name (e.g., "Python", "calculus", "TCP/IP")
• For concept-level knowledge: concept name (e.g., "Python list comprehensions", "recursion", "TCP handshake")
• For user-level learning preferences: "user"

Predicates:

• learning_outcome: What the user successfully understood or mastered
• struggle_area: Concepts where the user consistently makes errors or expresses confusion
• prerequisite_mastered: Foundational knowledge confirmed as solid (feeds into curriculum planning)
• learning_preference: User's stated or inferred preferences (e.g., "prefers code examples over theory")
• study_pattern: Observed patterns in how, when, or how much the user studies

Permanence levels:

• stable: Long-term transferable skills that persist across topics (e.g., "user has mastered recursion across languages")
• standard (default): Topic-specific knowledge in active study (e.g., "user knows Python list comprehensions")
• volatile: Temporary confusion, current struggle areas, or paused study states

Tags: Use tags like mastered, struggle, python, math, paused, preference, pattern

Example Facts

# From: user correctly answers quiz on recursion
memory_store_fact(
    subject="recursion",
    predicate="learning_outcome",
    content="user correctly explained base case, recursive case, and call stack behavior",
    permanence="stable",
    importance=8.0,
    tags=["recursion", "mastered", "fundamentals"],
    entity_id=<recursion_node_entity_id>  # from mind_map_node_get() or curriculum_next_node()
)

# From: user repeatedly struggles with closures
memory_store_fact(
    subject="Python closures",
    predicate="struggle_area",
    content="user confused about variable capture semantics in closures — mixes up early and late binding",
    permanence="volatile",
    importance=7.0,
    tags=["python", "closures", "struggle"],
    entity_id=<closures_node_entity_id>  # from mind_map_node_get() or curriculum_next_node()
)

# From: diagnostic — user already knows basic algebra
memory_store_fact(
    subject="algebra",
    predicate="prerequisite_mastered",
    content="user demonstrated solid understanding of algebraic manipulation and equation solving",
    permanence="standard",
    importance=7.0,
    tags=["math", "prerequisite", "algebra"],
    entity_id=<algebra_node_entity_id>  # from the relevant node dict
)

# From: user says "I prefer seeing code examples before theory"
# entity_id not required for user-level preference facts
memory_store_fact(
    subject="user",
    predicate="learning_preference",
    content="prefers concrete code examples before abstract theory",
    permanence=,
    importance=,
    tags=[, ]
)



memory_store_fact(
    subject=,
    predicate=,
    content=,
    permanence=,
    importance=,
    tags=[, ]
)