Documentation Index
Fetch the complete documentation index at: https://docs.superbryn.ai/docs/llms.txt
Use this file to discover all available pages before exploring further.
Integration methods
SuperBryn supports five ways to ingest production call data. The right method depends on your provider and use case.| Method | Best for | Real-time? |
|---|---|---|
| Webhook | Vapi, Retell, ElevenLabs, Bland | Yes |
| API key sync | Vapi, Retell, ElevenLabs, Bland | No (pull-based) |
| LiveKit SDK | LiveKit agents | Yes |
| Audio upload | Any provider, manual QA | No |
| Traces upload | Custom stacks with JSON logs | No |
Webhook
Webhooks push call data to SuperBryn in real-time as calls complete.- Go to Monitor and click Add Source
- Select your provider and choose Webhook
- SuperBryn generates a unique webhook URL
- Copy the URL and paste it into your provider’s webhook settings
- Calls will stream in automatically
API key sync
API key sync pulls call data from your provider on demand. Available for Vapi, Retell, ElevenLabs, and Bland.- Go to Monitor and click Add Source
- Select your provider and choose API Key
- Enter your provider’s API key (and optionally an agent ID to filter calls)
- Click Test Connection to verify
Sync modes
When syncing calls, you choose a scope:| Mode | Description |
|---|---|
| Incremental | Resume from where the last sync left off |
| Time range | Pull calls between specific start and end dates |
| Rolling window | Pull calls from the last N days |
| Recent N | Pull the most recent N calls (up to 10,000) |
LiveKit SDK
For LiveKit-based agents, SuperBryn integrates via thelivekit-evals Python package.
- Go to Monitor, open the Keys tab, and create a new API key
- Set the
SUPERBRYN_API_KEYenvironment variable in your agent’s environment - Install the SDK:
- Wire the webhook handler into your LiveKit agent code following the provided template
Audio upload
Upload recordings directly for analysis. Useful for spot-checking calls or analyzing recordings from providers without a direct integration.- Go to Monitor and click Upload
- Drag and drop audio files or click to browse
- Supported formats: MP3, WAV, M4A, OGG, FLAC
- Maximum 50 files per batch
- Optionally assign labels at upload time
Traces upload
Upload JSON trace files containing call records with transcripts and recording URLs.- Go to Monitor and click Upload Traces
- Select your JSON files (max 10 files, 50 MB combined)
- Choose the channel mode:
- Mono - single channel, requires speaker diarization
- Stereo - multichannel, agent and customer on separate tracks
- Optionally assign labels
Managing sources
All sources for an agent appear in the Monitor tab. You can:- Rename a source for clarity
- Enable/disable a source without deleting it
- Regenerate webhook tokens
- Delete a source and optionally its associated calls

