Agent Memory Systems
Design and evaluate short-term context, long-term vector memory, and CoALA-style memory types so agents retrieve the right facts instead of hoarding noise.
Overview
Agent Memory Systems is an agent skill most often used in Build (also Ship, Operate) that explains how to architect short- and long-term agent memory for reliable retrieval.
Install
npx skills add https://github.com/sickn33/antigravity-awesome-skills --skill agent-memory-systemsWhat is this skill?
- CoALA framing: semantic, episodic, and procedural memory types
- Seven principles including retrieval quality, chunking for retrieval, and memory decay
- Explicit anti-pattern: context isolation blocks effective memory
- Capabilities cover agent-memory and long-term-memory design
- Recommends testing retrieval accuracy and background memory formation
- 7 stated design principles for agent memory
- 3 CoALA memory types: semantic, episodic, procedural
Adoption & trust: 1.5k installs on skills.sh; 40.1k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
Your agent forgets everything between sessions because you treated memory as a dump of embeddings without retrieval design or the right cognitive memory types.
Who is it for?
Solo builders shipping persistent agents, copilots, or multi-session coding assistants who need a clear memory model before picking vector DBs.
Skip if: One-off chat tasks with no persistence needs, or teams wanting a turnkey hosted memory SaaS config with no architecture tradeoffs.
When should I use this skill?
You are architecting or debugging agent memory, context persistence, vector retrieval, or cognitive memory organization.
What do I get? / Deliverables
You leave with a CoALA-aligned memory plan—chunking, retrieval tests, decay policy, and typed stores—so agents recall the right context at the right time.
- Memory architecture outline (short-term vs long-term and memory types)
- Retrieval and chunking strategy with test plan before production
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Memory architecture is decided while you build agent tooling, before you scale sessions or ship memory-dependent features. Chunking, embeddings, retrieval, and memory-type choice are core agent-tooling design decisions, not generic frontend or SEO work.
Where it fits
Choose semantic vs episodic stores before wiring your first vector DB for a repo-aware agent.
Run retrieval accuracy checks on chunked docs before enabling memory in production.
Introduce decay rules when old episodic traces pollute answers on current sprint work.
Align procedural memory with onboarding flows so returning users skip re-explaining setup.
How it compares
Architecture and retrieval methodology—not a drop-in MCP memory server or a single-database integration skill.
Common Questions / FAQ
Who is agent-memory-systems for?
Indie developers and agent builders designing durable memory for LLM agents who care about retrieval accuracy and cognitive structure, not just bigger context windows.
When should I use agent-memory-systems?
During Build when designing agent tooling, before Ship when hardening retrieval tests, and during Operate when tuning decay and long-term store quality.
Is agent-memory-systems safe to install?
Review the Security Audits panel on this Prism page for the sickn33 catalog entry; the skill is conceptual guidance sourced from vibeship-spawner-skills (Apache 2.0).
SKILL.md
READMESKILL.md - Agent Memory Systems
# 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 - Context isolation is the enemy of memory - Right memory type for right information - Decay old memories - not everything should be forever - Test retrieval accuracy before production - Background memory formation beats real-time ## Capabilities - agent-memory - long-term-memory - short-term-memory - working-memory - episodic-memory - semantic-memory - procedural-memory - memory-retrieval - memory-formation - memory-decay ## Scope - vector-database-operations → data-engineer - rag-pipeline-architecture → llm-architect - embedding-model-selection → ml-engineer - knowledge-graph-design → knowledge-engineer ## Tooling ### Memory_frameworks - LangMem (LangChain) - When: LangGraph agents with persistent memory Note: Semantic, episodic, procedural memory types - MemGPT / Letta - When: Virtual context management, OS-style memory Note: Hierarchical memory tiers, automatic paging - Mem0 - When: User memory layer for personalization Note: Designed for user preferences and history ### Vector_stores - Pinecone - When: Managed, enterprise-scale (billions of vectors) Note: Best query performance, highest cost - Qdrant - When: Complex metadata filtering, open-source Note: Rust-based, excellent filtering - Weaviate - When: Hybrid search, knowledge graph features Note: GraphQL interface, good for relationships - ChromaDB - When: Prototyping, small/medium apps Note: Developer-friendly, ~20ms p50 at 100K vectors - pgvector - When: Already using PostgreSQL, simpler setup Note: Good for <1M vectors, familiar tooling ### Embedding_models - OpenAI text-embedding-3-large - When: Best quality, 3072 dimensions Note: $0.13/1M tokens - OpenAI text-embedding-3-small - When: Good balance, 1536 dimensions Note: $0.02/1M tokens, 5x cheaper - nomic-embed-text-v1.5 - When: Open-source, local deployment Note: 768 dimensions, good quality - all-MiniLM-L6-v2 - When: Lightweight, fast local embedding Note: 384 dimensions, lowest latency ## Patterns ### Memory Type Architecture Choosing the right memory type for different information **When to use**: Designing agent memory system # MEMORY TYPE ARCHITECTURE (CoALA Framework): """ Three memory types for different purposes: 1. Semantic Memory: Facts and knowledge - What you know about the world - User preferences, domain knowledge - Stored in profiles (structured) or collections (unstructured) 2. Episodic Memory: Experiences and events - What happened (timestamped events) - Past conversations, task outcomes - Used for learning from experience 3. Procedural Memory: How to do things - Rules, skills, workflows - Often implemented as few-shot examples - "How did I solve this before?" """ ## LangMem Implementation """ from langmem import MemoryStore from langgraph.graph import Stat