Skip to main content

Calling an Agent Inside a Step

Workflows can call agents to generate dynamic responses from within a step. This example shows how to define an agent, register it with the Mastra instance, and invoke it using .generate() inside a workflow step. The workflow takes a city name as input and returns a fact about the corresponding city.

Creating an agent​

Create a simple agent that returns facts about a city.

import { openai } from "@ai-sdk/openai";
import { Agent } from "@mastra/core/agent";

export const cityAgent = new Agent({
  name: "city-agent",
  description: "Create facts for a city",
  instructions: `Return an interesting fact based on the city provided`,
  model: openai("gpt-4o"),
});

Registering an agent​

To call an agent from a workflow, the agent must be registered in the Mastra instance.

import { Mastra } from "@mastra/core/mastra";
import { cityAgent } from "./agents/example-city-agent";

export const mastra = new Mastra({
  // ...
  agents: { cityAgent },
});

Calling an agent​

Get a reference to the registered agent using getAgent(), then call .generate() inside the step, passing in the input data.

import { createWorkflow, createStep } from "@mastra/core/workflows";
import { z } from "zod";

const step1 = createStep({
  id: "step-1",
  description: "passes value from input to agent",
  inputSchema: z.object({
    city: z.string(),
  }),
  outputSchema: z.object({
    facts: z.string(),
  }),
  execute: async ({ inputData, mastra }) => {
    const { city } = inputData;

    const agent = mastra.getAgent("cityAgent");
    const response = await agent.generate(
      `Create an interesting fact about ${city}`,
    );

    return {
      facts: response.text,
    };
  },
});

export const callAgent = createWorkflow({
  id: "agent-workflow",
  inputSchema: z.object({
    city: z.string(),
  }),
  outputSchema: z.object({
    facts: z.string(),
  }),
})
  .then(step1)
  .commit();