hoard sync
The hoard sync command triggers a sync from all configured data sources.
Usage
hoard sync [options]Options
| Option | Description |
|---|---|
--config <path> | Use alternate config file |
Examples
Sync All Sources
hoard syncSyncs all enabled connectors.
Output
$ hoard sync
Syncing local_files... Entities: 251, Chunks: 820, Tombstoned: 0, Errors: 0Syncing obsidian... Entities: 356, Chunks: 1325, Tombstoned: 2, Errors: 0Sync Process
For each enabled connector:
- Scan — Find all items in source
- Compare — Check content hashes for changes
- Process — Chunk new/modified content
- Index — Update database and FTS
- Tombstone — Mark missing items as deleted
Change Detection
Hoard uses content hashes to detect changes:
- Unchanged — Skip (fast)
- Modified — Re-chunk and re-index
- New — Process and add
- Deleted — Tombstone (soft delete, persists indefinitely)
Background Sync and Watcher
Hoard can sync automatically while the server is running (hoard serve or hoard mcp serve).
Configure in ~/.hoard/config.yaml:
sync: interval_minutes: 15 # 0 to disable scheduled sync watcher_enabled: false # Enable file watcher watcher_debounce_seconds: 2Connector Status
Check what’s been synced:
hoard connectors statusConnector Enabled Entities Last Synclocal_files yes 1,200 2026-02-03T10:15:00obsidian yes 1,256 2026-02-03T10:15:00notion_export yes 98 2026-02-03T10:15:00Troubleshooting
Sync Taking Too Long
- Reduce paths in config
- Narrow down
include_extensionsto only needed file types
Missing Content
- Check file extensions:
include_extensionsin config - Verify path exists
- Run
hoard doctorto diagnose
Sync Errors
Check the console output for detailed error messages for each file.
See Also
hoard add— Add data sourceshoard doctor— Diagnose issues- Connectors Guide — Configure connectors