What Agents Want from Your Product

Emmanuel Paraskakis

Founder, Level 250 | 3x VP, API Products

Your product was designed for humans. Your next million users won't be human.

1 month of Claude Pro included, so you can use Claude Code in the Workshop.

Your users are about to send agents to do their jobs for them. Agentic AI won't click through your UI, sit through your onboarding, or follow your navigation. It will look for an API, an MCP server, a CLI, or a skill. If it can't find one, it moves on.

Most products aren't ready for this. APIs built for human developers break when agents try to use them. Naming is vague. Errors don't help. Docs get skipped. Pagination fails. There's no MCP server, no CLI, and the only way in is through a UI that agents can't use.

In this workshop, you'll audit your own product through the eyes of an AI agent. You'll see live demos of what agents actually do when they hit your API. They skip your docs, guess from training data, follow descriptions over schemas, and retry from errors. Then you'll run the same experiments yourself.

Walk in not knowing if your product works for agents. Walk out with a 12-point agent-readiness checklist for your product, an API standards file you can share with your team, and a Skill you can install to design agent-ready APIs on your own.

What you’ll learn

You'll know how to test and improve your APIs for agent-readiness

  • Test if agents use documentation or specifications

  • Check for clear errors that help agents recover

  • Understand what agents expect from APIs

  • Know how to properly design an API that follows standards and best practices

  • Be able to validate OpenAPI specifications

  • Understand how to use and customize a Skill for API design

  • Use recomended approached to build and prioritize your own checklist

Workshop agenda

  • Optional: Introductions

    Join early to introduce yourself and meet participants in an informal setting before the workshop starts.

  • Agents are the new product persona

    The shift from clicking to delegating. The 4 things agents want (API, MCP, CLI, Skills), the one they don't (your UI), and why most existing APIs aren't good for agents.

  • How to make APIs usable for agents

    The 12+2 agent-readiness checklist with live demos for each insight. Watch agents skip docs, override schemas, and recover from errors.

  • Using APIs in Claude and checking usability

    Watch agents hit real APIs, then try it yourself. See where things go wrong, why they go wrong, and what to do about it.

  • What needs to change in your process and infrastructure

    Top 5 principles and the "Be Boring" rule. Design-first, standards, governance, evals. Auth, rate limiting, AI gateways, agent billing.

  • Write your agent-readiness checklist

    Exercise: pick your API, apply the 12+2 insights, write your checklist and draft your API standards as markdown. The /design-api skill and next steps.

  • Optional: Q&A and Next Steps

    Extended Q&A and help with prioritization. Path to further learning.

Learn directly from Emmanuel

Emmanuel Paraskakis

Emmanuel Paraskakis

3x VP Product (Apiary, Swagger, Oracle) | 1.3M APIs | Agent-ready APIs & MCP

Verisk
Moodys
Oracle
SmartBear
Precisely
See all products from Emmanuel

Who this workshop is for

  • API Product Managers who own APIs and need to know if agents can use them. Evaluate before you build.

  • Software Architects and Tech Leads responsible for API design and infrastructure. Prepare your platform for agent traffic.

  • Founders and Product Leaders hearing about agent-readiness from customers or competitors. Get a clear picture.

Prerequisites

  • A basic understanding of APIs and OpenAPI

    We'll test various APIs so you'll be ahead if you can understand these concepts

  • Familiarity with Claude Code or other codegen harnesses

    We'll supply a month of credits so you can complete the exercises and then move on to testing your own API products

What's included

Emmanuel Paraskakis

Live sessions

Learn directly from Emmanuel Paraskakis in a real-time, interactive format.

1 month of Claude Pro included so you can use Claude Code in the Workshop.

You'll receive your gift card before the workshop starts

API Design Skills included

Teach Agents to design standards-based, usable APIs with the included Skills.

Lifetime access

Go back to course content and recordings whenever you need to.

Community of peers

Stay accountable and share insights with like-minded professionals.

Certificate of completion

Share your new skills with your employer or on LinkedIn.

Maven Guarantee

Your purchase is backed by the Maven Guarantee.

Free resource

Why Agents Can't Use Your Product cover image

Why Agents Can't Use Your Product

Why your product is hard for agents to use

Agents are smart but they also seek the quickest cheapest path to an outcome. We'll show you where they fail.

Four things agents want and one they don't

Ideally you'll have one or more ways for agents to use tools and not have to scrape pages. We'll show what to use when.

How to check if your APIs are usable by agents

We'll show you how to come up with a list of prompts and desired/undesired outcomes so you can test your own product.

The top changes to make your APIs friendlier to agents

You'll take away a high-level plan and a 12-point checklist so you can get started towards agent-friendly API products.

Frequently asked questions

Maven for Teams

Reimbursement

Get your company to pay

Everything L&D needs: email template, receipts, and certificate of completion.

Get reimbursed

Team discount

Learn with your teammates

Save 20%+ when 2 or more teammates enroll in the same cohort.

Save 20%+ with a team

Private cohort

Run a cohort for your org

A dedicated cohort with a custom schedule and curriculum, tailored to your team.

Book a private cohort
Save 25% through Sunday

$499

USD

8 days left to enroll