Ask HN: How do you share agent context across a team?
The problem: that context is trapped on one machine. When he asks his agent something I already worked out with mine, he starts from zero. We're maintaining two "company brains" and they're drifting apart.
I don't want to share everything. Half my local context is personal or half-baked. I want a shared subset both agents read and write, with the rest kept private.
My plan: one `shared/` folder synced over Google Drive, both agents pointed at it. Company stuff in `shared/`, personal stays local. Simple, but I doubt it survives contact: sync conflicts when both write at once, no history of who changed what, no real boundary beyond the folder name.
How do you actually handle this?
- Shared folder, git repo, a database, something else? - How do you split shared vs. private without leaking one into the other? - What breaks when two agents write the same knowledge?