agent-memory-systems

Installation
Summary

Multi-layer memory architecture for agents: short-term context, long-term vector storage, and retrieval optimization.

  • Covers seven memory types: short-term (context window), long-term (vector stores), working, episodic, semantic, and procedural memory, each suited to different information patterns
  • Provides three core patterns: memory type selection, vector store choice, and chunking strategy to maximize retrieval accuracy
  • Highlights critical retrieval challenges: contextual chunking, metadata filtering, temporal scoring, and embedding model tracking to prevent "intelligence failures" caused by poor recall
  • Warns against common pitfalls: storing everything indefinitely, chunking without testing retrieval, and using a single memory type for all data
SKILL.md

Agent Memory Systems

Memory is the cornerstone of intelligent agents. Without it, every interaction starts from zero. This skill covers the architecture of agent memory: short-term (context window), long-term (vector stores), and the cognitive architectures that organize them.

Key insight: Memory isn't just storage - it's retrieval. A million stored facts mean nothing if you can't find the right one. Chunking, embedding, and retrieval strategies determine whether your agent remembers or forgets.

The field is fragmented with inconsistent terminology. We use the CoALA cognitive architecture framework: semantic memory (facts), episodic memory (experiences), and procedural memory (how-to knowledge).

Principles

  • Memory quality = retrieval quality, not storage quantity
  • Chunk for retrieval, not for storage
Related skills
Installs
1.4K
GitHub Stars
37.3K
First Seen
Jan 19, 2026