Handoffs between assistants
Stop in one AI. Pick up in another.
Save where a piece of work stands in one assistant and carry on from the next step in another, in either direction.
Cross-AI memory
ChatGPT remembers you in ChatGPT. Claude remembers you in Claude. RelayLink gives every assistant you connect the same memory of you, and the notes and handoffs you have saved, so the next one never starts from nothing.
What your assistants know about you visible to you
Illustrative · any line can be deleted from your account settings
The short answer
Not on their own: each keeps its memory inside its own app, and neither can read the other’s. What they can share is a third place. Connect both to your RelayLink account and switch memory on, and each one starts from the same memory of you, adds to it when you tell it something worth keeping, and searches it when a conversation needs more.
How it works
Add RelayLink to Claude, ChatGPT or another compatible assistant. One account works with as many of them as you use.
It stays off until you turn it on in your account settings. From then on your assistants keep what you ask them to remember, and, if you choose, the lasting facts they notice as you work, such as how you like to work or what you are building.
Each one reads what matters most when a session starts and searches the rest when the conversation calls for it, so a preference you mention to Claude on Monday is already known to ChatGPT on Tuesday.
Set it up in your AI, step by step
What you would say
The details
Questions
No, and nothing outside those apps can. Your connected assistants write to RelayLink when you ask them to remember something, or also as they work if you choose that, because they are asked to, so it works as far as each one follows the request.
No. It stays off until you switch it on in your account settings, and then an assistant keeps a fact only when you ask, unless you choose to let them keep what they notice as you work too. You can pause it at any time. Pausing stops new lines being added; what is already there stays until you delete it.
Yes. Every line is listed in your account settings with who recorded it, and each one has a delete button. You can also ask any connected assistant to forget a line, and that works even while memory is paused.
Theirs stays inside their app, which is fine if you only ever use one. This list is shared by every assistant you connect to your account, and you can read and edit all of it in one place.
You can, once. An import copies what one assistant knew on the day you moved. Shared memory keeps working while you use both, so a preference you mention to either one reaches the other.
No. Memory is your own record about yourself. It is never written into a message, and nobody you are in touch with can see it.
Up to 100 facts on the free plan and 1,000 on Pro, each up to 1,000 characters. A session starts with what matters most, what you told an assistant yourself and then the newest, kept to a fixed size so no conversation gets heavier as your memory grows, and your assistant searches the rest whenever the conversation needs it.
Claude and ChatGPT connect by signing in; ChatGPT needs its developer mode, on the web. Claude Code, Codex, VS Code and Zed sign in too, and other coding tools, such as Cursor and Gemini CLI, connect with a key from your account. The setup guide lists each one, what we have tested and when, and the apps that cannot connect yet.
It belongs to your account and is never shared. It is encrypted in transit and at rest. It is not end-to-end encrypted, because the service has to read it in order to hand it to your assistants.
Keep going
Create your account now and it is ready right away — nothing to wait for, nothing to redo.