Mastra

A Mastra agent runs on Rebuno unchanged apart from two seams. The model’s provider takes rebunoFetch, so every model call is recorded as an llm_call step, and tools run through defineTool, so every tool call is recorded as a tool_call step. On a re-dispatch the agent runs from the top and recorded steps replay instead of calling the model or the tool again.

Install

npm install rebuno @mastra/core @ai-sdk/openai zod

Model calls

Mastra takes AI SDK providers, which accept a custom fetch. Pass rebunoFetch:

import { createOpenAI } from "@ai-sdk/openai";
import { rebunoFetch } from "rebuno";

const openai = createOpenAI({ fetch: rebunoFetch });

See LLM calls.

Tools

Declare the Rebuno side with defineTool, and pass it as the execute of a Mastra tool:

import { createTool } from "@mastra/core/tools";
import { defineTool } from "rebuno";
import { z } from "zod";

const sendEmail = createTool({
  id: "send_email",
  description: "Email the support summary.",
  inputSchema: z.object({ body: z.string() }),
  execute: defineTool({
    name: "send_email",
    idempotency: "at_most_once",
    execute: async ({ body }: { body: string }) => mail.send("ops@acme.com", body),
  }),
});

Mark anything with a side effect, such as sending an email or creating a ticket, at_most_once. See idempotency.

The agent

Mastra’s Agent and Rebuno’s Agent share a name, so import one under an alias:

import { Agent as MastraAgent } from "@mastra/core/agent";
import { Agent } from "rebuno";

async function process(input: { query: string }) {
  const openai = createOpenAI({ fetch: rebunoFetch });
  const assistant = new MastraAgent({
    id: "support",
    name: "Support",
    instructions: "Investigate customer issues and email support summaries.",
    model: openai("gpt-5.5"),
    tools: { send_email: sendEmail },
  });
  const { text } = await assistant.generate(input.query, { maxSteps: 12 });
  return { answer: text };
}

const agent = new Agent("support");
await agent.serve({ port: 5000 }, process);

What Rebuno adds

Policy. Every model and tool call is checked against policy before it runs. A denied tool returns the rule’s reason to the model as the tool result, so the agent can take a different path.
Approvals. A tool that requires approval parks the execution. Once it’s approved, the agent is dispatched again, the earlier steps replay, and the approved call runs.
Recovery. If the worker dies partway through, the next dispatch replays every completed step and continues from the first one that didn’t finish. Model calls that already ran are not paid for twice.

Full example

examples/frameworks/typescript/mastra_agent.ts is a support agent that investigates a customer issue, creates a ticket, and emails a summary after approval.

Get started

Writing agents

Frameworks

Reference

Install

Model calls

Tools

The agent

What Rebuno adds

Full example

​Install

​Model calls

​Tools

​The agent

​What Rebuno adds

​Full example

Install

Model calls

Tools

The agent

What Rebuno adds

Full example