How do I share memory across my team's Claude Code sessions?
By Soumen Sarkar · October 2026
Short answer: Connect every teammate's Claude Code to one shared memory the sessions read and write as they work: what's being changed in each part of the code, what was decided, and what was assumed. CLAUDE.md holds the stable rules; the shared memory holds what changes by the hour.
Why doesn't each session already know?
Each Claude Code session starts from the code, its CLAUDE.md files and, on one machine, its own auto memory. That's plenty for one person. On a team, the facts that matter most aren't in any of those:
- that a teammate's session is reshaping the checkout right now, and what it will look like when it lands;
- that the team decided this morning to store money in whole pence;
- that someone's agent assumed last night that partial refunds don't return shipping, and a product manager corrected it at 8am.
Claude Code's own memory features work across one person's sessions. Nothing in a session tells it what another person's sessions are doing.
What does shared team memory need to do?
Four things, without anyone copying anything anywhere:
- Brief a session before it changes an area. The first time a session reads or edits a file in a part of the code, it gets that area's work in progress (with its plan), the decisions that apply, and what was tried and dropped. See area brief.
- Record calls as they're made. Decisions and assumptions go into the shared record the moment a session makes them, so the next session in that area follows them. See assumption.
- Bring answers back mid-work. When someone confirms or corrects an assumption, or work a session depended on lands, the session hears about it at its next turn.
- Leave something behind. When a session stops, a handoff says what's done, what's left and the next step.
How do you set it up?
Keep stable instructions (commands, conventions, boundaries) in CLAUDE.md, and put the live context in one place every session reaches. Claude Code can reach a shared source through its hooks (which run at session start and as files are read and edited) and through an MCP server (which gives the agent tools to record and look things up).
Campfyrr is that shared source, packaged as a Claude Code plugin. Install it with one token, and every teammate's sessions start with the team's brief, get area briefs as they work, and record their decisions, assumptions and handoffs. It never stores your code: only file paths and what happened. For the whole set-up, see shared team memory for Claude Code, and for the split between files and live memory, what goes in CLAUDE.md, AGENTS.md and team memory.
Words used here
- Area brief: An area brief is what a person or agent is told the first time they touch a part of the code: what's changing there now, what was decided, and what was tried.
- Work in progress: Work in progress is work under way, shown with its plan, so others build on it rather than around it.
- Assumption: An assumption is the best available answer to an unknown, stated and acted on, kept visible so anyone can correct it.
- Team record: The team record is everything a team has written down that a person or agent can search before asking: decisions, rules, handoffs, assumptions and the reasons behind changes.
The principle behind this: Answer it yourself: Look first. If the answer isn't there, assume it or decide it.
Move as one at agent speed. Nobody waits. Add Campfyrr to Claude: it briefs every agent before it touches the code, keeps assumptions and handoffs where the team can see them, and never takes your code (what it keeps).