liveBook · Manning

Semantic memory

Semantic memory is long-term memory for meanings, concepts, and general knowledge organized by meaning rather than by specific personal episodes.

How 4 Manning books explain it

AI Agents in Action, Second Edition

Semantic memory is memory retrieved according to the meaning of stored information rather than only its explicit graph relationships or literal terms.

How it is used
Deploys semantic memory as long-term agent storage for retrieving facts, concepts, and contextual information by meaning.
How deep it goes
Explains encoding, indexing, retrieval, augmentation, backend selection, and hybrid search, while deferring production operations to separate work.
Examples
Uses vector-store retrieval, generated retrieval questions, and a ChromaDB-plus-knowledge-graph hybrid agent.
What you can do after
Operate/configure semantic retrieval and combine it with graph search for richer agent memory, while recognizing ephemeral-demo limitations.

Read how AI Agents in Action, Second Edition explains semantic memory →

AI Agents in Action

Semantic memory is a memory form organized around meanings, including conceptual information and associated feelings or emotions.

How it is used
Stores and retrieves relevant facts and concepts for agent long-term memory, and supports meaning-based memory retrieval.
How deep it goes
Explains semantic processing, query extraction, and Nexus store setup, but notes that specific memory function prompts cannot be configured.
Examples
Compares how the same statements populate two memory types and describes adding facts and preferences to a Nexus semantic store.
What you can do after
Apply it via Nexus by creating a semantic memory store and adding facts or preferences for semantic categorization.

Read how AI Agents in Action explains semantic memory →

Context Engineering

Semantic memory is durable factual, conceptual, and stable knowledge that remains consistent across contexts.

How it is used
Personalizes agent behavior by supplying stable user or domain knowledge through context engineering.
How deep it goes
Defines semantic memory and gives examples, but specifies no semantic-specific retrieval algorithm.
Examples
Uses a Python-preferring data scientist, a premium subscriber, and a significant yield-curve inversion.
What you can do after
Understand how stable facts can inform agent personalization and adaptation.

Read how Context Engineering explains semantic memory →

Designing AI Agents

Semantic memory is persistent memory for general facts, concepts, and their relationships rather than specific events.

How it is used
Stores general facts, concepts, rules, and procedures as an agent-memory category distinct from active context and specific experiences.
How deep it goes
Defines semantic memory by its contents and places it in a three-tier analogy, without detailing its storage implementation.
Examples
Names facts, rules, and procedures as semantic-memory contents.
What you can do after
Understand semantic memory’s role alongside working memory and episodic memory.

Read how Designing AI Agents explains semantic memory →

Also covered in

What to know first

Related concepts

Questions about semantic memory

What is semantic memory?

Semantic memory is long-term memory for meanings, concepts, and general knowledge organized by meaning rather than by specific personal episodes.

Which Manning books explain semantic memory?

AI Agents in Action, Second Edition, AI Agents in Action, Context Engineering and Designing AI Agents.

What should I know before learning semantic memory?

Manning books that teach semantic memory usually introduce vector database, context window and dialogue first.

What is semantic memory a kind of?

Semantic memory is a kind of memory. Memory is a storage capability or mechanism for retaining information or computational state for later access or use.

What is semantic memory part of?

Semantic memory is part of AI agent. An AI agent is a computational system that uses artificial intelligence to interpret information and select or perform actions toward goals while interacting with an environment.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.