agent-memory-systems
Memory is the cornerstone of intelligent agents. Without it, every interaction starts from zero. This skill covers the architecture of agent cognitive architectures that organize them."
Profile →npx skills add sickn33/agentic-awesome-skills --skill agent-memory-systems --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Outlines an architecture for agent memory, including short-term/context memory, long-term/vector-based memory, and cognitive-type memories (semantic, episodic, procedural). It emphasizes retrieval quality, chunking for retrieval, context isolation avoidance, and decay strategies. It provides capabilities like semantic, episodic, procedural memory and related retrieval/formation/decay operations.
How it works
The skill presents a reference framework (CoALA) for organizing memory into semantic, episodic, and procedural layers. It lists memory types, the associated tooling and vector stores, and embedding-model options. It includes concrete implementation sketches for LangMem-based memory usage (semantic.upsert, episodic.add, procedural.add) and contextual runtime retrieval (semantic.get, episodic.search, procedural.search). It offers patterns for vector store selection (Pinecone, Qdrant, Weaviate, ChromaDB, pgvector) and describes chunking strategies (Fixed-Size, Semantic, Structure-Aware, Contextual, Code-Specific). It advocates Background Memory Formation (LangGraph-based background processing) and Memory Consolidation/Decay with sample code for time-based decay and utility-based pruning. It demonstrates a Memory Retrieval at Runtime flow to assemble context from profile, past experiences, and skills.
When to use it
Triggered when designing or refining an agent memory system that requires organized, retrievable knowledge across short-term and long-term scopes. It is applicable when choosing memory types, vector stores, embedding models, and chunking strategies to optimize retrieval quality.
What it can touch
- Memory frameworks like LangMem
- Vector stores: Pinecone, Qdrant, Weaviate, ChromaDB, pgvector
- Embedding models: OpenAI embedding variants, nomic-embed-text-v1.5, all-MiniLM-L6-v2
- Runtime memory operations: semantic.upsert, episodic.add, procedural.add, semantic.get, episodic.search, procedural.search
- Background processing and consolidation routines as examples
Caveats
- The content is a conceptual and code-ahead blueprint with numerous placeholder snippets and architectural guidance; actual integration requires adapting to specific agent runtime and environment.
- Licensing notes appear as MIT for the skill; usage should comply with compatible licenses of included tooling and models.
- Some sections present representative code blocks without guaranteeing production readiness or completeness, and rely on external libraries not fully specified here.
# 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 - memor
- Principles
- Capabilities
- Scope
- Tooling
- Memoryframeworks
- Vectorstores
- Embeddingmodels
- Patterns
- Memory Type Architecture
- LangMem Implementation
- Memory Retrieval at Runtime
- Vector Store Selection Pattern
- Pinecone (Enterprise Scale)
- Qdrant (Complex Filtering)
What does the agent-memory-systems skill do?
Memory is the cornerstone of intelligent agents. Without it, every interaction starts from zero. This skill covers the architecture of agent cognitive architectures that organize them."
How do I install it?
Run `npx skills add sickn33/agentic-awesome-skills --skill agent-memory-systems --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From sickn33/agentic-awesome-skills, a repository with 44,414 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.