skills.rest
bahayonghang
bahayonghangCOMMUNITY
Explore all skills by this author →
GitHub

memory-system

Index Markdown notes into a local SQLite vector/full-text index.

16|1|Updated Jan 2, 2026
Search Tags:#markdown#indexing#memory#sqlite#notes#semantic-search#vector-search
One-click install
npx skills add https://github.com/bahayonghang/my-claude-code-settings --skill memory-system
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-system
Source: https://github.com/bahayonghang/my-claude-code-settings/tree/main/skills/memory-system
Command: npx skills add https://github.com/bahayonghang/my-claude-code-settings --skill memory-system

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sentence-transformers, numpy, and includes scripts (resource) and references (resource) components.

What problem does it solve?

The Memory System indexes Markdown notes into a local SQLite vector/full-text database to enable fast cross-session recall and querying across conversations.

Core Features & Use Cases

  • Incremental indexing: detects changes with SHA256 and only indexes updated files for efficiency.
  • Hybrid search: combines vector embeddings with full-text search for accurate results.
  • Memory add/cleanup: supports adding new notes and pruning old ones to maintain a lean index.
  • Use case: organize your personal knowledge base of Markdown notes and retrieve relevant passages across sessions with a single query.

Quick Start

Index your Markdown notes: python3 memory.py index --dir memory Search your memory: python3 memory.py search "Your query" --top 5 Add memory: python3 memory.py add "Key insight" --dir memory --file note.md Check status: python3 memory.py status --dir memory

Frequently Asked Questions about memory-system

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I index Markdown notes for semantic search across sessions?▼

To index Markdown notes for semantic search, the system detects file changes using SHA256 hashes and incrementally adds updated notes to a local SQLite database, creating a searchable vector and full-text index for cross-session retrieval.

How does hybrid search work for personal knowledge management?▼

Hybrid search for personal knowledge management combines vector embeddings generated by sentence-transformers with SQLite full-text search, retrieving accurate and contextually relevant Markdown note passages based on your query.

Do I need Python and sentence-transformers to build a local memory index?▼

Yes, you need Python and the sentence-transformers library to generate the vector embeddings required to build and query the local SQLite memory index. Numpy is also used as an optional dependency to enhance search performance.

What is the best way to search Markdown notes across multiple conversations?▼

The best way to search Markdown notes across conversations is querying a local SQLite index using the command line, which applies hybrid vector and full-text search to return the top relevant passages from your knowledge base.

Can I add new notes and clean up old entries in a SQLite memory index?▼

Yes, you can add new notes to the SQLite memory index and clean up old entries. The system supports adding specific memory entries via command line and pruning outdated files to maintain a lean and efficient searchable database.

Related Skills

Explore More Education & Research Skills→
anderskevanderskev
20

sqlite-vec

Store and query vector embeddings in SQLite using vec0 virtual tables.

Community
Intermediate
megalithicmegalithic
223

brave-search

Query the Brave Search API and convert page content to Markdown.

Community
Intermediate
eddmanneddmann
2

zettelkasten

Store and retrieve atomic notes with semantic connections in SQLite.

Community
Advanced
davpatel605-beepdavpatel605-beep

pinecone

Manage scalable vector database storage and retrieval for AI applications.

Community
Advanced