Updated July 2026

Building Agents with Claude

Class Duration

7 hours of live training delivered over 1-2 days to accommodate your scheduling needs.

Student Prerequisites

Comfort with everyday office software: documents, spreadsheets, email, and a web browser
Basic experience chatting with an AI assistant is helpful but not required
No technical background or programming experience required

Target Audience

Business users and team leads who want to build their own custom AI assistants with Claude: operations, HR, finance, marketing, sales, legal, support, and management. Ideal for teams standardizing on Claude who want reusable, purpose-built assistants grounded in their own documents and ways of working. No technical background is assumed, and nothing in the course requires one.

Description

This practical, hands-on course teaches business users how to build their own custom assistants with Claude, no programming required. Participants learn how Claude works, then build reusable assistants using Projects that combine custom instructions with a body of trusted knowledge. The course covers writing instructions that shape how an assistant behaves, adding uploaded documents as knowledge, packaging repeatable know-how as Skills, and using connectors so an assistant can work with current information and other tools. Testing, verification, and responsible use receive close attention: refining an assistant until it is dependable, checking its output, protecting sensitive data, and sharing an assistant with a team. Every topic is practiced against realistic workplace scenarios so participants leave able to build, refine, and share useful assistants right away.

Learning Outcomes

Explain in plain language how Claude works and where it is strong or weak.
Create a reusable assistant with a Claude Project that combines instructions and knowledge.
Write clear custom instructions that shape an assistant's role, tone, and scope.
Add uploaded documents as trusted knowledge and keep them current.
Package repeatable know-how as a Skill an assistant can reuse.
Use connectors and tools so an assistant can reach current information and other systems.
Test and refine an assistant, verifying its answers against its sources.
Protect sensitive and confidential information when building and sharing assistants.
Share an assistant with a team and establish shared practices for using it.

Training Materials

Comprehensive courseware is distributed online at the start of class. All students receive a downloadable MP4 recording of the training.

Software Requirements

Students need access to Claude through a Claude account or their organization's Claude for Work plan, and a web browser. No installation required.

Training Topics

Understanding Claude as an Assistant

What Claude is and how it generates responses, in plain language
What Claude does well: language, reasoning, and long documents
Where Claude falls short: facts, arithmetic, and recent events
Chat versus a reusable, purpose-built assistant
Everyday scenarios where a custom assistant helps

Building with Projects

What a Project is and when to create one
Combining custom instructions with a shared body of knowledge
Organizing work so a team draws on the same assistant
Naming and describing a Project so others understand it
Previewing and using a Project in daily work

Writing Custom Instructions

The anatomy of clear instructions for an assistant
Defining the assistant's role, scope, and boundaries
Setting tone, audience, and response format
Handling questions outside the assistant's purpose
Refining instructions based on how the assistant responds

Adding Knowledge

Uploading documents as trusted knowledge
Choosing which sources belong in an assistant
Keeping knowledge accurate and up to date
How grounding reduces made-up answers
Checking answers against the underlying documents

Packaging Know-How as Skills

What a Skill is and how it captures repeatable work
Turning a proven process into a reusable Skill
Combining Skills with instructions and knowledge
When a Skill is a better fit than a one-off prompt
Maintaining Skills as your processes change

Connectors and Tools

How connectors let an assistant reach beyond a conversation
Using web search for current information
Connecting approved data sources and tools
Understanding what a connector can access on your behalf
Reviewing results before relying on them

Testing, Verification, and Responsible Use

Trying an assistant against realistic questions
Recognizing made-up answers and checking sources
Matching the depth of checking to the stakes of the task
Protecting customer, employee, and company data
Working within your organization's AI and data policies

Sharing an assistant with a team
Gathering feedback from the people who use it
Keeping instructions, knowledge, and Skills maintained
Building a shared library of assistants that work
Staying current as Claude's capabilities change