diff --git a/README.md b/README.md index 423c01d..2343b0d 100644 --- a/README.md +++ b/README.md @@ -1,14 +1,16 @@ -# BearDrive — a synced file system for AI agents +# BearDrive — Google Drive for AI agents **BearDrive** mounts any folder as a synced volume: its contents stay synchronized across all your devices through cloud object storage, every change is tracked (who, when, on which device), and everything keeps working offline. The CLI is `bdrive`. -It is built for AI agent workflows — give your agents on every machine the -same `~/agent-workspace`, and notes, plans, memory files, and artifacts -follow them everywhere, with a full audit trail of which agent or human -changed what. +Two things it's for: **sharing files with people** — any synced file +becomes a public URL that renders as a page — and **sharing context across +AI agents**: give every agent on the team the same folder as memory, and +your agent knows what their agent knows. Notes, plans, findings, and +artifacts follow the team everywhere, with a full audit trail of which +agent or human changed what. ```console $ bdrive login # once per device (browser sign-in) diff --git a/website/index.html b/website/index.html new file mode 100644 index 0000000..08f6963 --- /dev/null +++ b/website/index.html @@ -0,0 +1,581 @@ + + +
+ + +One synced folder for your whole team — humans and agents. + Share any file with people as a public URL; give every agent the same + memory, synced across machines in seconds. Real files. No server required.
+Works with the storage you already have
+HOW IT WORKS
+BearDrive mounts any folder as a synced volume. Your agents just + write files — every editor, every tool, every agent already knows how to do that.
+Init the folder. bdrive init connects the project and
+ starts a background daemon. Teammates run the same command and converge on
+ the same files within seconds.
Agents write, everyone sees — including their agents. Claude writes
+ wiki/report.md on your machine; moments later it's in every
+ teammate's folder, where their own agent reads it — attributed, versioned,
+ kept forever.
Share by URL. bdrive share wiki/report.md prints a
+ public link that renders the file as a page — always the latest synced
+ version, no account needed, revocable any time.
WHY BEARDRIVE
+Share files with people. Anything in the folder is a
+ bdrive share away from being a public URL that renders as a page —
+ markdown, HTML, PDFs. No accounts for viewers, no "can you re-send that?",
+ always the latest version, revocable any time. Dropbox-link simplicity for
+ everything your team produces.
Share context across agents. The folder is shared + memory. What your agent learns today — findings, plans, decisions, notes — + is on your teammate's disk in seconds, so their agent starts tomorrow's + session already knowing it. Context stops being siloed per laptop; + every agent on the team reads from, and writes to, the same brain.
+FEATURES
+Sync is table stakes. What agent teams actually need is + provenance, shareability, and a security model that assumes files + are written by software.
+Real files, zero lock-in. Not a virtual filesystem, not an API. A + normal folder on disk — grep it, open it in your editor, point any agent at + it. Stop syncing any time; the files stay.
+Every change tracked, forever. Who changed each file, when, from + which device — every past version viewable and downloadable. An agent + overwriting your doc is an annoyance, not a loss.
+Public share links. Any synced file becomes a rendered page at an + unguessable URL: HTML as a page, markdown Obsidian-style, PDFs inline. + Sandboxed, rate-limited, living until you revoke it.
+Teams & organizations. Projects belong to your org; only + members see or sync them. Invite a teammate with one expiring link. + Accounts, device registry, and history come built in.
+Offline-first, conflict-safe. Append-only journals and + deterministic replay: devices converge without a coordination server, and + concurrent edits keep both versions as conflict copies.
+Bring your own storage. S3, GCS, R2, MinIO, or a plain shared + directory — BearDrive needs nothing but dumb object storage. Or sync + through a hub and devices never hold cloud credentials at all.
+CLAUDE CODE
+wiki/q3-findings.md
+ (synced to the team) and here's a public link for anyone outside it:wiki/q3-findings.md — synced to this machine
+ four minutes ago — churn concentrated in self-serve accounts, and…GET BEARDRIVE
+The entire product is open source under AGPL-3.0 — server, teams, + history, share links, nothing held back. The cloud exists for teams who'd + rather never think about buckets.
+bdrive web is the viewer, uploads, and sync hubbdrive login and gobrew install, bdrive init, done. Files travel to people as URLs; context travels to every agent on the team.
+Same layout, one variable changed. Scroll — the verdict under each card is the honest trade-off.
+Real files. No server required.
+ +Real files. No server required.
+ +Real files. No server required.
+ +Real files. No server required.
+ +Real files. No server required.
+ +