The outcome

Learn the interaction loop using a small task with a clear outcome.

Mastra is a Open-source TypeScript framework for AI agents, typed tools, graph workflows, memory, retrieval, evaluation, and observability. Building TypeScript agents for open-ended tasks and typed, resumable workflows for controlled multistep processes, with shared memory, tracing, and evaluation. This guide narrows that broad capability into one repeatable outcome, with checkpoints that keep the source material and your judgment in the loop.

Before you begin

Set the boundary before the tool starts.

Choose one real task, identify who will use the result, and decide what evidence or test will make the result acceptable. Gather only the source material needed for that task. If the work contains confidential, personal, regulated, or client-owned information, confirm that the platform and account are approved before sharing it.

Troiana principle

AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.

Step by step

A workflow you can repeat.

  1. 01

    Define the user task, autonomy boundary, data class, model, tools, output schema, memory, workflow states, approvals, quality, safety, latency, cost, and rollback requirements.

  2. 02

    Start with one narrowly scoped agent for open-ended reasoning or a deterministic workflow when steps and control flow are known, registering only the minimum typed tools.

  3. 03

    Run locally in Studio with server-side secrets, representative and adversarial cases, bounded steps and tokens, observable tool calls, validated outputs, and no automatic consequential actions.

  4. 04

    Review traces, eval scores, permissions, memory isolation, failures, resume behavior, latency, and cost, then pin dependencies, canary deployment, and retain a tested fallback.

Working standard

What good use looks like.

  • State the outcome before the background.
  • Provide the real source material.
  • Review the result before expanding the task.

Typed schemas validate structure, not truth, authorization, or safe intent. Agents can loop or misuse tools, and workflows can replay or resume with stale state. Treat prompts, memory, MCP content, tool arguments, and resume payloads as untrusted; authorize outside the model, make side effects idempotent, cap requests and cost, isolate tenants, protect snapshots and traces, and require approval for consequential actions.

Official references

Check the current product documentation.

Features, plan limits, availability, and data controls change. These official pages are the starting points used for this collection.