Workshop

Introduction to Claude Agents for User Researchers

Nikki Anderson

AI infrastructure for user research

Build Claude agents that help support research with your frameworks

Level: Built for researchers who have already put together several Claude skills.

I built this workshop for researchers who have hit the wall, where one skill handles a single task well, but a whole stretch of your process still lives across separate chats. Maybe you have a skill that drafts a research plan and another that writes up insights, and you keep opening a fresh chat for each step, copying context between them, doing the coordination yourself.

An agent changes that. It calls on the skills you already have, and handles the coordination between them, planning from a raw stakeholder request through to a plan ready to send, or synthesis from raw notes through to insights, in one flow. I will show you what makes an agent different from a skill, how to tell which part of your process is ready to hand over, how to scope an agent so it stays focused instead of trying to run your job at once, and I will walk you through real agents I have already built, showing you my own process end to end so you can mimic it for your own work. You leave knowing how to build an agent and more comfortable and confident to build one yourself

What you’ll learn

Gather the skills an agent calls on and write the handoff instructions, so a whole phase of your process runs up to your standard

  • See the real difference between a skill that handles one task and an agent that calls on several

  • Know where agents live and who can see them across chat, Cowork, and Projects

  • Walk through one of my agents in detail to understand what it is

  • Use the same mechanical versus judgment test that scopes a skill, now at the scale of a whole phase

  • Tell a process that is ready to hand over from one that still needs you in the middle

  • Learn why 8 to 15 skills tends to be the range that keeps an agent focused

  • Understand how to scope out an appropriate amount of tasks for one agent to handle

  • Learn how to incorporate subagents to run simultaneously

  • Walk through what I decide before I open Claude, the process, the audience, the skills it calls on

  • See the instructions that tell an agent when to hand off from one skill to the next

  • Understand how to incorporate MCPs into your agents

  • Know how to test agents across recent work to spot any issues

  • Understand how to think from a stakeholder's POV when creating an agent

Workshop agenda

  • What a Claude agent is, and how it differs from a skill

    See the line between a skill that handles one task and an agent that calls on several, chains skills together, and moves through a full piece of your process from request to finished plan

  • Which part of your process is ready to hand over

    Use the same mechanical versus judgment test that scopes a skill, now at the scale of a whole phase, and see the signs a process is ready to become an agent

  • Scoping an agent so it stays on track

    Learn why one agent should run one piece of your process, why 8 to 15 skills tends to be the range that works, and what happens when an agent tries to do too much

  • How I put an agent together

    Walk through what I decide before I open Claude, how I gather the skills an agent calls on, and the instructions that tell it when to hand off between them

  • Reading, testing, and iterating on an agent

    See how I read through what Claude built before I trust it, test it against recent human made work, and why an agent usually takes a few more rounds than a single skill

Learn directly from Nikki

Nikki Anderson

Nikki Anderson

Contact

Helping user researchers feel more confident using AI

See all products from Nikki

Who this workshop is for

  • User researchers who have built several skills and are looking to chain them together into a flow

  • Teams who want consistent agents that generate effective output each time, no matter who starts it

  • Teams who are looking to democratize parts of the research process to stakeholders but want to uphold quality

What's included

Nikki Anderson

Live sessions

Learn directly from Nikki Anderson in a real-time, interactive format.

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.

A bonus agent starter kit to help you build your first agent

Get started more quickly with your next UXR agent with this starter kit

Maven Guarantee

Your purchase is backed by the Maven Guarantee.

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

$250

USD

Oct 27
·

10:30am–1:30pm EDT

Enroll