Install
openclaw skills install omi-integrationSync recordings from Omi AI wearables (Omi, Limitless, etc.) via API and webhooks. Auto-sync transcripts, process recordings, and organize by device/date.
openclaw skills install omi-integrationSync and manage recordings from Omi AI wearables (Omi, Limitless pendant, etc.).
mkdir -p ~/.config/omi
echo "YOUR_API_KEY" > ~/.config/omi/api_key
chmod 600 ~/.config/omi/api_key
echo "https://api.omi.me" > ~/.config/omi/backend_url
# Or for self-hosted:
echo "https://your-backend.com" > ~/.config/omi/backend_url
omi-sync
omi-sync --days 7
omi-list
omi-get <recording-id>
cat webhook-payload.json | omi-webhook-handler
Recordings are stored in:
~/omi_recordings/
├── YYYY-MM-DD/
│ ├── <recording-id>/
│ │ ├── metadata.json
│ │ ├── transcript.txt
│ │ ├── audio.wav (if available)
│ │ └── summary.md
└── index.json
Configure your Omi app to send webhooks to your endpoint:
recording.created, transcript.updatedThe skill includes a handler (omi-webhook-handler) that processes real-time events.
Recordings are automatically tagged by device:
{
"recording_id": "rec_123",
"device_id": "limitless-001",
"device_name": "Limitless Pendant",
"device_type": "wearable",
"context": "work",
"transcript": "Meeting notes...",
"created_at": "2026-02-02T15:38:00Z"
}
Base URL: https://api.omi.me/v1 (configurable)
GET /recordings - List all recordingsGET /recordings/:id - Get recording detailsGET /recordings/:id/transcript - Get transcriptGET /recordings/:id/summary - Get AI summaryPOST /webhooks/register - Register webhook endpointAuto-sync every hour:
0 * * * * /path/to/omi-sync --days 1 >> ~/.local/share/omi/sync.log 2>&1
Or use Clawdbot cron for integrated scheduling.