Contact Modulate to obtain ToxMod API credentials and access to developer documentation at console.modulate.ai
Integrate ToxMod with your existing VoIP provider (Vivox, Agora, or Discord Social SDK) using the provided SDK; for Discord Social SDK integration a single configuration call routes player audio to ToxMod
Configure moderation categories (harassment, hate speech, grooming, threats) and set your alert thresholds and automated action triggers in the ToxMod dashboard
Set up webhooks or polling to receive moderation events from ToxMod; each event includes a confidence score, detected category, and a clipped audio reference for human review
Implement a human review queue for borderline-confidence events before applying permanent bans; reserve automated instant action for high-confidence severe violations
Known gotchas
ToxMod processes audio server-side; player audio must be routed through your VoIP infrastructure to ToxMod — client-side capture is not a supported integration path
Latency of voice moderation decisions means ToxMod is not designed to prevent the first few seconds of harmful speech; it is built for pattern detection and consequence workflows, not real-time content blocking
Consult your legal team before storing audio clips that ToxMod references for review; audio retention policies vary by jurisdiction and must be disclosed in your game's privacy policy
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp