Developer Documentation & API Usage

API Integration

Voiceflow exposes a flexible REST API that allows developers to embed and control AI agents across any interface or platform. Agents can be invoked programmatically, passing user inputs and receiving structured responses for rendering in custom UIs.

POST https://general-runtime.voiceflow.com/state/user/{userID}/interact
Authorization: VF.DM.xxxxxxxx
Content-Type: application/json

{
  "action": {
    "type": "text",
    "payload": "I need help with my order"
  }
}

Key Platform Capabilities

Agentic Context Engine: Manages stateful, multi-turn conversations with enriched context passing between workflow nodes.
LLM Flexibility: Integrate GPT-4, Claude, Gemini, or BYOM via model configuration settings — no vendor lock-in.
Environment Pipelines: Separate dev, staging, and production environments allow safe iteration and controlled rollouts without agent rebuilds.
Observability & Evals: LLM-powered evaluation tools provide automated quality scoring, intent coverage analysis, and conversation-level drill-downs at scale.
Webhooks & Integrations: Connect to external APIs, CRMs, ticketing systems, and databases using built-in HTTP request blocks or custom code functions.
Performance: 500ms latency SLA for voice agents; 300K messages/minute throughput for enterprise deployments.