Route tasks between assistants — and hand off to a human when confidence is low

The scenario assesses each incoming task, judges its complexity and its own confidence, hands it to the right agent, and emails devino when confidence is low.

  • When in doubt it goes to a human, not to a guess
  • Simple tasks don't burn a heavyweight assistant
  • The confidence threshold is tuned to your risk appetite
  • Every decision comes with a confidence score
Webhook

How it works

To connect devino and Webhook, you don't need a developer: a ready-made scenario links them in minutes.

  1. Starts when: Webhook
  2. Then: Workflow Configuration
  3. Then: Supervisor Agent
  4. Check: Check Confidence Score
  5. If yes: Execute Selected Agent
  6. If no: Send Fallback Alert
  7. Then: Routing Decision Parser
  8. Then: OpenAI Model - Supervisor
  9. Then: Simple Task Agent Tool
  10. Then: Complex Task Agent Tool
  11. Then: OpenAI Model - Complex Agent
  12. Then: OpenAI Model - Executor
  13. Then: OpenAI Model - Simple Agent(mini-model)

You can launch this devino + Webhook integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.