0.2.4
Vault Rooms 0.2.4 — Live cursors and safer collaboration
Highlights
- Added live cursors and note presence for CRDT-enabled Markdown notes.
- See teammates’ authenticated display names, carets, and selections while editing the same note.
- Presence colors are assigned by the relay and remain unique among active users in the room session.
- Colors adapt to Obsidian light and dark themes.
- Presence is ephemeral and note-scoped: it disappears when the editor, session, or connection closes.
- V1 intentionally does not include a participant bar.
Sync reliability
- Fixed pinned WebSocket disconnects that could leave pending requests stranded.
- Prevented rejected updates from recreating documents at their old path after a rename.
- Improved CRDT recovery across rename, reconnect, unmount, and remount flows.
- Fixed editor rebinding after session resync.
- Fixed failed editor bindings being cached as rejected promises.
- Preserved exactly-once bidirectional editing while live cursors are active.
Security and protocol hardening
- Prevented identity changes on established WebSocket connections.
- Same-device re-authentication can no longer downgrade CRDT or presence capabilities.
- Invalid re-authentication tokens are still fully authenticated and rejected.
- Improved safe, user-facing validation errors while keeping raw network failures in diagnostic logs.
LAN sharing and diagnostics
- Improved LAN address validation and warnings for loopback, unspecified, stale, and link-local addresses.
- Expanded the live-editing diagnostics command with presence transport, publication, visible-peer, and bound-editor status.
- Added clearer troubleshooting messages for invite, connection, permission, and cursor failures.
Compatibility
- Both devices need Vault Rooms 0.2.4 or newer to exchange live cursors.
- The room must have CRDT live editing enabled.
- Both users must have the same Markdown note open and retain
file:readpermission for that path. - Minimum supported Obsidian version remains 1.12.7.
Full changelog: 0.2.3...0.2.4