Shared Context: Notes and Docs the Whole Workspace Can Use
Chat memory lives in one person's thread. Shared notes and docs are the source of truth every agent and teammate can see.
By the flAim team
Chat memory in a single-model product is one person talking to one model. In flAim the thread is shared: @ another agent and it continues that conversation. Shared context is the notes and docs layered on top — every agent, every teammate, even on a new send.
If you have ever forwarded a ChatGPT transcript and watched the next person re-explain the project, you have felt the difference. The transcript was not the brief. It was a private artifact of one thread.
Thread memory vs workspace context
| Conversation memory | Shared notes and docs | |
|---|---|---|
| Who sees it | Usually one person, one product account — unless it is a flAim workspace thread | Everyone invited to the workspace |
| Who uses it | The model in that thread | Every agent you add, on Chat or Build |
| What happens on a new chat | You paste again, or you lose the constraint | The brief is still in the room |
| How you share | Export, screenshot, “Claude said…” | Invite. Live drafts, presence, same agents |
The collaboration primitive is the room
Sharing a workspace is not a bolt-on “team chat.” It is the same session: agents, thread, notes. Prompting stops being a private craft and becomes something you can correct in public. Bad prompts get fixed. Good ones get reused.
Limits follow the product: Free is one workspace. Pro, Power, and Team unlock unlimited shared workspaces. Personal plans spend the owner’s credits when a collaborator runs a managed chat. Team plans use a pooled wallet. ZDR routing is a Team data-handling option, not a substitute for putting the brief in notes.
How to attach files and notes: Documentation. For a concrete setup, see the research workspace and writing studio playbooks.
Put the brief where agents can see it
Open a workspace, add Notes, paste the job once. Then invite someone instead of screenshotting.
Related: Shared context in docs · Research playbook · Definition
We're building this in the open and we read everything. If you try it and something's clumsy, tell us. We'd rather hear it from you than not hear it.