Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.xysq.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
What is xysq?

xysq gives your AI agents persistent, identity-aware memory — with you in control of what they remember. Connect xysq once, and every tool you use (Claude, Cursor, Codex, and more) shares the same memory: facts you’ve taught it, files you’ve uploaded, decisions your team has made. No more starting from zero in every new conversation. The boundary is consent. Your memory belongs to you. Agents only retain and recall within the limits you set.

​
Get started in four steps

1

Connect your agent

Install the xysq CLI and pair it with the AI agent you use. The CLI wires up the xysq MCP server and installs a small skill that teaches your agent the retain/recall protocol.

CLI quickstart

One command to install, authenticate, and connect every agent on your machine.

Claude Desktop

Manual setup for Claude Desktop if you prefer it over the CLI.
2

Organise — upload your first file

Drop a document, a chat transcript, or a link into your xysq vault. xysq reads it, extracts knowledge, and makes it available to every connected agent.

Organise

How vaults, folders, and uploads work.
3

Chat — ask about that file

Open the xysq chat (or any connected agent) and ask a question about what you just uploaded. The agent recalls relevant snippets automatically — no manual context needed.

Chat

How chat works, how it scopes memory, and what the agent can do.
4

Teams — create your first team for collaborative memory

Invite a teammate, and your team shares a vault. Decisions, context, and uploads accumulate as one memory the whole team can recall from — across every agent each member uses.

Teams

How team vaults, roles, and shared memory work.

​
Need a walkthrough?

If anything’s unclear, book a 30-minute call and we’ll walk through your setup together.