Перейти к содержимому
TakoAPI
Назад к навыкам

aister-vector-memory

от alekhm

Search & Research
635Число загрузок по данным clawskills.sh
1Просмотры
0 лайковВойдите, чтобы поставить лайк навыку
—Ещё нет оценок
Ваша оценка

Кратко

Vector memory for Aister — search by meaning, not by grep!

Описание

Что делает этот навык

Adds semantic search to Aister's memory using PostgreSQL with pgvector and the e5-large-v2 embedding model. Queries match by meaning rather than exact keywords. Supports both English and Russian.

Finds relevant memory entries even when the query shares no words with the stored text, unlike grep or keyword search.

Когда его использовать

  • Finding past notes when you remember the topic but not the exact wording
  • Searching stored preferences and settings by description
  • Recalling what was written about a project without knowing precise terms
  • Looking up identity or behavior notes by concept
  • Retrieving context from memory files before starting a new conversation

Пример сценария

ВВОД

User asks: /search_memory my communication style

АГЕНТ
  1. 1

    Converts the query to a 1024-dimension vector using e5-large-v2

  2. 2

    Runs cosine similarity search against stored vectors in PostgreSQL

  3. 3

    Filters results above the 0.5 similarity threshold

  4. 4

    Returns up to 5 matching memory chunks with source context

ВЫВОД

Excerpts from MEMORY.md or USER.md describing the user's communication preferences, ranked by semantic similarity

Требования

  • PostgreSQL 16 with pgvector extension (requires root or sudo to install)
  • VECTOR_MEMORY_DB_PASSWORD environment variable set to the PostgreSQL user password

Установка

clawhub install alekhm/aister-vector-memory

Выполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.

Ссылки