GitHub Copilot memory you can read on your phone

Copilot in VS Code has a memory tool. It keeps notes in three scopes: about you across all your workspaces, about one repository, and about one conversation. All three live in VS Code’s own storage on the computer, none of them goes into the repository, and nothing in the docs says they sync.

Its memory tool keeps notes on the computer it runs on, so connect Copilot to Pando and have it write decisions into your outline, which every computer reaches, and so does every other agent you allow to read it.

Set it up

VS Code runs the sign-in itself: it registers with Pando and opens your browser once. Nothing is pasted anywhere.

What each place is for

• .github/copilot-instructions.md or AGENTS.md, for instructions that belong to the repository and the team.
• Copilot’s memory, for recall on this computer.
• A branch of your outline, for decisions and their reasons that have to survive a second computer, a second agent, or a question you ask from your phone.

Before you connect it at work

On Copilot Business and Enterprise, MCP servers stay off until an admin enables the MCP servers in Copilot policy. GitHub’s docs: "The policy is disabled by default." On a personal plan there is no such step.

Working with other agents

Claude Code, Codex and ChatGPT connect to the same outline. What Copilot wrote down in the office is what Claude Code reads at home, and you read both on your phone in between.

What it costs

Nothing, for 1,000 bullets, with agents included and no card. Pando does not meter agent calls.

Deeper

• Codex memory that follows you to the next machine
• Two agents, one memory

Connect an agent, about two minutes · The twelve tools · Which note apps an agent can reach · Who runs this