TL;DR
78 terms covering the memory architecture of LLM agents: working/short/long-term/episodic/semantic/procedural memory · RAG (retriever, ranker, reranker, chunking, hybrid search, query decomposition, fusion decoding, GraphRAG) · vector stores (HNSW, IVF, ANN, pgvector, Qdrant, Weaviate, Milvus, Pinecone) · knowledge graphs (entity, relation, triple store, property graph, Cypher, SPARQL) · agent-memory patterns (MemGPT/Letta, Mem0, MemoryBank, A-Mem, Generative Agents, Voyager, Reflexion) · context engineering (context window, prompt caching, lost-in-the-middle, NIAH, compaction) · forgetting strategies (decay, supersession, eviction, freshness floor).
Each entry has a primary-source citation and a date accessed. Live search, category filter, and A–Z jump nav below. Flagship terms link to the deep glossary for longer narrative entries.
Sibling references on agentsbooks.com
- Pillar essay: Agent memory and knowledge — the canonical home for this vocabulary on the AgentsBooks hub.
- Pillar essay: Context graph, explained — sibling pillar pairing for the knowledge-graph and context-graph terms above.
- Worldview: Anatomy of a Firm — the 8 primitives every service firm needs (Memory and Knowledge are two of them).
- Compare with: the deep agentic glossary — longer narrative entries for the flagship terms (RAG, Memory, Knowledge, Context graph).