Core Concepts

Understanding how Hebbrix thinks about memory will help you use it better. It's simpler than it sounds.

01. Memories

A "memory" is just a piece of information. It could be:

A chat message ("User prefers dark mode")
A document you uploaded (your company's FAQ)
A preference ("Speaks Spanish")
Literally anything you want the AI to remember

02. Collections

Collections are folders for organizing memories. Like:

customer-support - All support conversations
product-docs - Your documentation
user-preferences - Settings and choices

When searching, you can filter by collection to find specific types of memories faster.

03. The 3-Tier Memory System

Hebbrix mimics how human memory works with three layers:

Short-Term Memory (STM)

High priority, fades quickly. Recent stuff - like things from today or this week.

Example: "User asked about pricing 2 hours ago"

Medium-Term Memory (MTM)

Refreshed if accessed again. Things you remember for weeks or months.

Example: "User's name is John, prefers email communication"

Long-Term Memory (LTM)

Permanent knowledge. Important facts that should never be forgotten.

Example: "Our refund policy is 30 days", "John is a premium customer"

Hebbrix automatically manages these tiers using the Ebbinghaus Forgetting Curve. Memories decay over time unless they're accessed again, just like in real brains.

04. Knowledge Graph

Behind the scenes, Hebbrix builds a graph of how things connect. For example:

Knowledge graph

John → works_at → Acme Corp
John → prefers → email
Acme Corp → industry → SaaS
email → faster_than → phone_calls

This helps the AI understand relationships and make smarter connections. You don't need to manage this. It happens automatically.