Learn

What is a brain?

A KBrain brain is a packaged, queryable knowledge source that lets people and AI agents access an expert's methods, style, and know-how - on demand, from any AI assistant.

See brains in action

Explore use cases

A brain is a packaged, queryable knowledge source built from an expert's real materials - documents, examples, methods, decisions, and voice. Once built, it can be queried by a person or an AI agent through any compatible AI assistant. The brain does not chat. It answers from its source material.

The plain-English definition

Think of a brain as a curated library that your AI assistant can search on demand. You choose what goes in it. The AI retrieves from it. The result is answers grounded in a specific expert's knowledge - not in the averaged output of a general model.

What it contains - 01

What can be inside a brain

A brain is not a chatbot and not a database. It is a curated knowledge layer.

📝

Methods and workflows

Step-by-step processes, decision frameworks, repeatable playbooks. The how-to knowledge that experts build over years.

💡

Examples and cases

Real examples, annotated outputs, and before-and-after cases. The source material that makes guidance concrete.

🗂️

Documents and decisions

PDFs, notes, decision logs, and institutional records. The written record of what an expert knows and why they made certain choices.

A brain is only as good as what goes into it. The more specific and practitioner-grade the source material, the more useful the brain.

How a brain differs from a chatbot

The Difference - 02

Brain vs. chatbot

🤖A chatbot

· Generates responses from model training data

· No specific source - cannot be traced or updated

· Locked inside one interface or app

· Built and maintained by a developer

· Answers from averages, not from a specific expert

1 app

accessible only
inside that interface

🧠A KBrain brain

→ Retrieves from curated, traceable source material

→ Every answer attributed to its source

→ Works via MCP in Claude, ChatGPT, Cursor, and more

→ Built by any expert - no code required

→ Answers from a specific person's knowledge

∞

any MCP client
any AI assistant

A chatbot is a product. A brain is a knowledge layer - it works inside the AI assistants you already use.

Who uses brains

Creators and experts: package your voice, style, and methods so AI assistants can write and advise like you
Teams: capture senior expertise and make it queryable for new hires and AI agents
AI agent builders: give agents a trusted, permission-scoped knowledge source to query at runtime
Learners and practitioners: access expert brains from the marketplace to get grounded guidance on demand

A brain does not replace an AI assistant. It gives your AI assistant the domain-specific, expert-curated context it needs to answer well. The brain is the knowledge; the assistant is the interface.

See brains in action

Browse the marketplace to see published brains, or create your own and connect it to your AI assistant.

Explore use cases

Frequently asked questions

Do I need to know how to code to build a brain?

No. You upload your documents, PDFs, and source material through the KBrain interface. KBrain handles indexing and makes the brain queryable. No code required.

Can a brain be kept private?

Yes. Brains are private by default. You choose who can access them. You can share a brain with specific users or teams, or publish it to the public marketplace.

How does a brain get queried?

A brain is accessible via MCP - the Model Context Protocol. Connect your KBrain account to Claude, ChatGPT, or any compatible AI assistant. When you ask a question, the assistant queries the relevant brain and grounds its answer in the brain's source material.

What is the difference between a brain and a RAG system?

A RAG system is a developer-built pipeline that retrieves documents to augment a model. A KBrain brain is a no-code, managed knowledge layer with permissions, source attribution, and MCP connectivity built in. You focus on the knowledge; KBrain handles the infrastructure.