memory-taxonomy

name	memory-taxonomy
description	Lifestyle domain memory classification — subject/predicate taxonomy, permanence levels, tags, and example fact patterns for storing taste, consumption, and hobby knowledge
version	1.0.0
tools_required	["memory_store_fact","memory_recall","memory_search"]

Lifestyle Memory Taxonomy Skill

Purpose

Load this skill when classifying and storing a taste preference, consumption note, hobby, or Spotify-enriched fact. It defines the subject/predicate pairs, permanence levels, tags, and worked example calls for memory_store_fact().

Lifestyle Domain Taxonomy

Use memory_store_fact() with the following subject/predicate pairs:

Subject: Use "user" for all personal preference facts. Use "spotify:artist:{id}" or "spotify:playlist:{id}" for Spotify-enriched facts.

Taste Preference Predicates (stable permanence):

• likes_genre: music genre preferences and dislikes
• likes_artist: favourite artists or acts
• likes_cuisine: cuisine types the user enjoys
• favorite_restaurant: preferred dining spots and why
• favorite_recipe: beloved recipes or dishes
• hobby: active hobbies and leisure interests
• food_preference: dietary patterns, ingredient preferences
• food_dislike: foods to avoid (allergies, aversions, dislikes)
• routine: daily routine patterns (morning rituals, evening wind-downs, focus modes)

Current Consumption State Predicates (volatile permanence):

• watches: currently watching (TV shows, films)
• reads: currently reading (books, articles, comics)
• plays: currently playing (video games, board games)
• listens_to: current listening focus (album, artist rotation, playlist)

Spotify-Enriched Predicates (stable permanence, Spotify subjects) (listed below as Subject Type | Predicate; pass only the predicate after the | to memory_store_fact(predicate=...)):

• spotify:artist:{id} | listening_pattern: rotation intensity and frequency over time
• spotify:playlist:{id} | purpose: what the playlist is for (focus, commute, party, etc.)
• spotify:playlist:{id} | context: when/where/why the playlist is used

Permanence levels (these map to the memory_store_fact(permanence=...) parameter):

• stable: Stable preferences that persist, such as cuisine tastes, genre opinions, favourite artists, hobbies, and routines
• volatile: Temporal state, covering what's currently being watched, read, played, or listened to

Tags: Use tags like music, food, restaurant, cuisine, tv, film, book, game, hobby, routine, artist, genre, spotify, dislike, allergy

Example Facts

# User mentions loving a genre
memory_store_fact(
    subject="user",
    predicate="likes_genre",
    content="loves jazz — especially 1960s modal jazz and contemporary jazz-fusion",
    permanence="stable",
    importance=8.0,
    tags=["music", "genre", "jazz"]
)

# User names a favourite artist
memory_store_fact(
    subject="user",
    predicate="likes_artist",
    content="Bill Evans — cites his trio recordings as some of their favourite music",
    permanence="stable",
    importance=8.0,
    tags=["music", "artist", "jazz"]
)

# User mentions a food they avoid
memory_store_fact(
    subject="user",
    predicate="food_dislike",
    content="dislikes coriander / cilantro — strong aversion, not just mild preference",
    permanence="stable",
    importance=9.0,
    tags=["food", "dislike", "allergy-adjacent"]
)

# User mentions a favourite restaurant
memory_store_fact(
    subject="user",
    predicate="favorite_restaurant",
    content="Koya — favourite udon spot, love the cold noodle dishes in summer",
    permanence="stable",
    importance=7.0,
    tags=["restaurant", "japanese", "udon"]
)

# User mentions what they're currently watching
memory_store_fact(
    subject="user",
    predicate="watches",
    content="currently watching Severance S2 — halfway through, loving it",
    permanence=,
    importance=,
    tags=[, , ]
)


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