liveBook · Manning

Memory bank methodology

As explained in Designing AI Agents

A structured practice of maintaining project context in a collection of persistent Markdown files for agent sessions.

The Memory Bank methodology preserves project continuity across agent sessions with a maintained set of files. The agent reads the files at session start, replacing context lost when a new session begins. section 4.1.3

The maintained files capture distinct kinds of project context:

Before adding a dedicated memory layer, check which authoritative project sources already preserve what later work needs. Extra persistence adds a control surface and is justified when needs such as automatic extraction, isolation, updating, recall, or cross-system reuse warrant it. section 4.1.3

How Designing AI Agents covers it

How deep it goes
Names five maintained file categories and their session-start role, but does not specify how to update or retrieve them.
Examples
Uses project, context, progress, pattern, and technology files as the persistent context set.
What you can do after
Operate an agent across sessions using maintained project files as its startup context.
How it connects
Connects agent memory with persistent context and filesystem-based storage.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Designing AI Agents. The section links point to the parts of the book it draws on.