Key Concepts
This is the "how does this thing actually work?" section. Not a wall of jargon. Not a system design interview. Just the mental models you need to understand what's happening under the hood so you can get the most out of your assistant.
The WorkspaceYour assistant's brain is a folder. Here's what's in it.
Tools & SkillsWhat your assistant can do, and how it learns new tricks.
Memory & ContextHow memory organizes and recalls what your assistant knows.
Self-improving SkillsHow completed work becomes a reusable procedure.
Model ProfilesPick which LLM runs each job, and override it when you need to.
ChannelsWhere you can talk to your assistant.
SchedulingHow your assistant gets things done on its own clock.
Web SearchHow your assistant looks things up online, and which provider does it.
OAuth IntegrationsConnect your assistant to Slack, Gmail, GitHub, Notion, and more.
You don't have to read these in order. Jump to whatever's interesting.