OpenCode with Memory Bridge

OpenCode with memory that stays when you switch models

Connect OpenCode to Memory Bridge once and it reads from, and writes to, the memory of your work. What you save here is there in your other tools.

Say it in Claude, carry on in OpenCode

Claude
You

Branches are named by work type then ticket number. Commit messages are in English, in the imperative.

Claude

Saved.

Saved to memory · Branch convention
One memory
OpenCode
You

Create a branch and commit this fix.

With Memory Bridge

Created branch fix/1482 and committed with an imperative message, following your saved branch convention.

From memory · Branch convention

Illustrative example

What does OpenCode remember on its own?

OpenCode reads an instructions file in the project and lets you switch models whenever you like. The new model does not know what you told the last one.

We reviewed this description in October 2026. These tools change often, so check the official help page for current details.

What does Memory Bridge add?

Memory that does not belong to one model. Switch as often as you want: your project context sits in one place, read by whichever model you choose.

Connect OpenCode in three steps

  1. 1

    Add Memory Bridge

    Add Memory Bridge to the OpenCode config file under external services, then sign in.

  2. 2

    Sign in and pick your projects

    Choose which projects it can reach, and whether it reads only or reads and writes.

  3. 3

    Carry on working

    Ask OpenCode as you always do. It finds what your request needs in your memory.

Common questions

Does Memory Bridge replace OpenCode's own memory?

No. Your project instruction files stay where they are and keep doing their job. Memory Bridge adds what does not fit in those files and makes it available to your other tools.

Do I need a paid OpenCode plan?

Connecting outside services in OpenCode may be limited to certain plans or versions. Check the tool's official help page.

What can OpenCode see of my memory?

Only the projects you allowed, with the permission you chose. You can disconnect it at any time.

What if the connection fails?

We have tested this connection and it works. If you hit a problem, write to us through the support channel with the error message.

Keep reading

Memory for coding agents

Instruction files tell a coding agent how to write code. How to give coding tools a memory of decisions and reasons that lasts across sessions and tools.

Cursor

Rules say how you write code. Connect Cursor to Memory Bridge and stop re-explaining your work.

GitHub Copilot

Connect GitHub Copilot to Memory Bridge and it reads from the memory of your work. What you save in GitHub Copilot is there in your other AI tools.

One memory. Every tool you use.

The trial is invite-only for now. Request access and we will write to you when your seat opens.