Get the risk score, risk explanation, biometric checks, and device checks for a specific session ID. The user_logs field is also included for backward compatibility but is deprecated — use the /v1/sessions/{sessionId}/events endpoint to retrieve a session’s event log.
Documentation Index
Fetch the complete documentation index at: https://docs.roundtable.ai/llms.txt
Use this file to discover all available pages before exploring further.
Use your secret API key as the bearer token
The unique session identifier (from window.getRoundtableSessionId())
Session data retrieved successfully
Unique identifier for the session
"abc123def456"
User identifier provided during script initialization (if any)
"user-456"
Tags associated with this session
["intake-survey", "checkout-flow"]Overall risk score for the session (0-100). Higher scores indicate higher risk.
0 <= x <= 10075
Short explanation of the factors contributing to the risk score
"User was flagged for multiple likely bot behavior, indicating a medium to high risk of fraud."
Suggested action based on the risk score
Auto-accept, Accept with monitoring, Manual review or require additional verification, Auto-reject or high-priority review, Unknown "Manual review or require additional verification"
Deprecated. Chronological list of user actions recorded during the session. This field will be removed soon — use the /v1/sessions/{sessionId}/events endpoint instead.
[
{
"action": "Signed in",
"user_time": "Mar 25, 2025 14:30:45",
"unix_timestamp": 1743055845000
},
{
"action": "Navigated to /dashboard",
"user_time": "Mar 25, 2025 14:30:50",
"unix_timestamp": 1743055850000
},
{
"action": "Updated credit card",
"user_time": "Mar 25, 2025 14:32:15",
"unix_timestamp": 1743055935000
}
]Results of biometric analysis for bot detection
{
"agent_behavior": "Not detected",
"programmatic_typing": "Detected",
"teleporting_mouse": "Not detected",
"no_corrections": "Detected",
"all_pasted": "Unknown",
"jump_scrolling": "Detected",
"centered_clicks": "Detected"
}Results of device and network checks for environment anomalies
{
"bot": "Detected",
"virtual_machine": "Not detected",
"software_renderer": "Not detected",
"vpn": "Detected",
"tor": "Not detected",
"location_spoofing": "Unknown"
}