Overview
What Stateful is, how the pieces fit together, and where to start.
Stateful is a persistent memory layer that sits above every AI you use. It stores decisions, corrections, preferences, and constraints as typed facts — then surfaces the right slice of that knowledge to whatever AI you're working with, whether that's Claude, ChatGPT, Cursor, or your own agent.
The problem it solves
Every AI session starts from zero. You explain the same context, re-state the same preferences, and re-teach the same corrections — across every new chat, every model switch, every agent restart. You become the integration layer between AI and your own knowledge.
Stateful removes that burden. Connect your sources once. Your context stays current, persists across resets, and is readable by any AI that can call an MCP server or accept a context injection.
How the pieces fit together
| Piece | What it does | Availability |
|---|---|---|
| Wiki | Your hub — a structured, editable profile of who you are | Live |
| Connectors | Pull in data from email, calendar, GitHub, Slack, and more | Live |
| MCP server | Exposes your context to any AI that speaks the Model Context Protocol | Live |
| Claude Code hook | Adds your context to every Claude Code prompt, and lets an agent watch the session | Live |
| Browser extension | Pulls your context into Claude.ai and ChatGPT as you prompt | Closed beta |
Where to start
- Connect your sources — Google, GitHub, Slack, whatever you use most. This is what Stateful builds your profile from.
- Browse your wiki — see what Stateful knows about you and fill in the gaps.
- Connect via MCP — let any MCP-capable AI (Claude.ai, Cursor, and more) pull your context on demand.
New here? Getting Started walks through all of this end to end.
The browser extension is still in closed beta. If you're a beta user and want it, reach out and we'll get you set up.