AI Digital Brain Glossary

Plain-English definitions for the terms behind AI-powered knowledge. Digital brain is the term we use at BrainGraph; second brain is the term you have probably heard. Both are defined here — and every definition shows how the pieces connect.

Digital brain
A living knowledge graph built from your real work — transcripts, documents, conversations, and frameworks — that AI agents can reason with, not just search. Where a notes archive stores what you know, a digital brain connects it and answers questions grounded in your own data. It is the term BrainGraph uses for the step beyond a second brain.See how BrainGraph builds your digital brain →
Second brain
The popular term, from the Personal Knowledge Management world, for an external system that stores what you know — usually notes in a tool like Obsidian or Notion. A second brain stores; a digital brain connects and reasons. BrainGraph builds what people often call a second brain, then takes it further into a digital brain an AI can actually act on.From a second brain to your own digital brain system →
Knowledge graph
A way of storing information as entities — people, projects, topics — connected by relationships, often pictured as dots and lines. Unlike a folder of documents, a knowledge graph lets you see how everything links and ask questions across all of it at once. It is the structure underneath a digital brain.BrainGraph →
Graph database
The type of database a knowledge graph runs on. It stores not just data but the connections between data — which is what lets a digital brain answer questions a normal document store or spreadsheet cannot. It shows how information is linked, not just what the information is.
Node
A single point in a knowledge graph — a person, an organization, a project, a topic, or a concept. Nodes are the things your digital brain knows about; edges are how they connect.
Edge
A relationship between two nodes in a knowledge graph — for example, "this meeting discussed pricing" or "this client uses that framework." Edges are what turn a pile of separate facts into a graph you can reason across.
Entity
A real thing your digital brain knows about — a person, company, project, or concept — with a single canonical record. Matching every mention in your documents back to the right entity is what keeps the graph clean and trustworthy.
Fact
A single statement in your digital brain that connects two entities — the atomic unit of knowledge. Facts carry their source, so you can always trace where a piece of knowledge came from.
Knowledge base
A store of what you or your business knows. A knowledge base is a database; a digital brain is a knowledge base structured as a graph, so an AI can reason across it instead of only retrieving from it.
AI agent
A piece of software that can think and act within your knowledge space — prepare a meeting, score a lead, draft a message — using your digital brain as its source of truth. An agent decides nothing on its own authority; it surfaces and drafts, and you keep the judgment.Connect agents to your brain via the Claude connector →
Agent orchestration
Coordinating multiple AI agents so they work together on top of one digital brain — each handling a part of a larger workflow, all reasoning from the same shared knowledge.
Skill file
A markdown file — often with a .skill extension — that gives an AI agent step-by-step instructions for one specific task or workflow. The AI reads it and follows it; a good skill file behaves like a small, dedicated brain for a narrow job.
Soul file
A file that defines an AI agent's behavior and character — how it responds. Give an agent a "strict" soul file and its output comes back strict. Where a memory file is what the agent knows, a soul file is how it says it.
Memory file
A file that holds an AI agent's knowledge — the short, core layer of what it knows. Deeper knowledge lives in separate files; the memory file is the concise reference the agent reads to give grounded answers. Paired with a soul file, it is what makes an agent's responses both informed and in-character.
MCP (Model Context Protocol)
An open standard that lets an AI tool — Claude, ChatGPT, and others — securely connect to an outside source, like your digital brain, and use it during a conversation. BrainGraph ships an MCP connector so you can query your brain and save notes straight from chat.The BrainGraph MCP connector →
RAG (Retrieval-Augmented Generation)
A technique where an AI retrieves relevant information from an outside source before it answers, instead of relying only on its training. A digital brain is that source — and a graph-based digital brain retrieves connected context, not just matching snippets, so the answer reflects how your knowledge actually fits together.
Personal Knowledge Management (PKM)
The practice of capturing, organizing, and reusing what you learn. PKM is the discipline behind the "second brain" — and a digital brain is PKM taken one step further: organized so an AI can act on it, not just so you can find it later.
Ingestion
The process of turning your raw material — transcripts, emails, documents, article URLs, videos — into a structured knowledge graph. Ingestion is the step that makes a digital brain out of a pile of files.

Ready to build your own digital brain?

Start free with 40 documents — turn your transcripts, notes, and frameworks into a knowledge graph your AI can reason with.