Leapforce Academy Getting Started with Leapforce Introduction
View transcript
Hi, and welcome to Leapforce.
If you’re new to AI agents, this might feel like a lot — new tools, new terms, and an idea you’re not sure how to turn into something real.
That’s exactly why we built this course. You don’t need to become an AI expert, and you don’t need to know how to code. We’re going to keep it simple and learn by doing.
So, what is Leapforce? Leapforce helps you build AI agents that take care of real work for you. Think of an agent as an AI teammate — you give it a job, teach it how you want that job done, connect the tools and information it needs, and let it get to work.
You could build an agent that reviews customer feedback, drafts a weekly report, answers common questions, or organizes information for your team. Unlike a simple chatbot, an agent doesn’t just answer — it can follow steps, use your tools and data, pinpoint problems, recommend actions, and execute on your approval.
In this course, you’ll learn to build your first agent from a template so you don’t have to start from scratch. You’ll then customize it to fit your workflow, teach it with skills, test it, and publish it for real use.
By the end, you won’t just understand what an agent is — you’ll have built one.
So don’t worry about learning everything at once. Pick one real task you want AI to help with, and let’s start building.
Let’s get started.
Introduction
Overview
AI sounds exciting. But getting started can feel overwhelming.
There are so many tools, features, and new AI terms to learn. You may know there are things you could automate, but you are not sure where to start or how to turn an idea into something that actually works.
That is exactly why we created this course.
You do not need to become an AI expert. You do not need to know how to code. You just need to learn the basics, build something real, and learn by doing.
In this course, we will take you from your first idea to your first working agent. You will learn how to build it, teach it, automate it, and improve it.
By the end, you will know enough to start using Leapforce for your real work.
What you’ll learn
You will learn how to:
- Build an AI agent for a real task you want to automate.
- Teach your agent with skills and a knowledge base.
- Test and publish your agent, and share it with your team.
Before you start
Prepare: One recurring work task you could describe in a few sentences. Avoid sensitive or irreversible work for your first build.
Completion checkpoint
Write one sentence: “My agent helps [audience] achieve [outcome] by [task], while a human reviews [decision or action].”
View transcript
Welcome back.
Last lesson, you picked one real task you want AI to help with. Now let’s turn that into your first agent — and you don’t have to build it from scratch.
Leapforce gives you over 100 ready-made agent templates, covering common jobs across every part of the business. Every template comes with a dashboard, automation, and report already built in.
Start by typing what you need in the search bar, or browsing the template library, and picking one close to your real task. It doesn’t need to be a perfect match — just the closest starting point.
Next, connect your tools and data. This is what brings the dashboard, automation, and report to life — without it, your agent has nothing real to work with.
Now for the part that makes this fast: customizing your agent. You do it just by chatting with it — the same way you’d ask an assistant for a report. Tell it what you need, it does the work, and you approve before anything gets done. No forms, no setup screens, just a conversation.
Give your agent a name your team will recognize, then run a test with a real sample input.
Don’t expect perfection on the first try — that’s normal, and it’s what the rest of this course is for. Right now, you just want a working first draft.
Take a look at the output — the dashboard, the automation, the report. Does it do roughly what you wanted? Good — that’s your first agent.
Next, we’ll teach it to do that job even better, using skills and a knowledge base.
Launch Your First Agent (Using Templates)
Overview
You don’t have to build from scratch. Leapforce gives you 100+ ready-made agent templates to choose from, covering common jobs across every part of the business. Every template comes with a prebuilt dashboard, automation, and report already set up — you just need to connect your tools and data so they can actually pull in your real information. From there, you customize the agent simply by chatting with it — the same way you’d ask an assistant for a report, then approve it to get things done.
In this lesson, you’ll pick a template close to your real task, connect your tools and data, customize it through chat, and run your first test. This is the moment your idea stops being an idea.
What you’ll learn
- How to choose the right template that fits your task.
- What comes prebuilt with every template — dashboard, automation, and report.
- How to connect your tools and data so they actually work.
- How to customize your agent, and get things done quickly just by chatting with it directly.
- How to run your first test, and optimize for the best output.
Before you start
Bring the task you described in the Introduction. Have it ready in one or two sentences — you’ll paste it straight into your agent’s instructions.
Completion checkpoint
Write one sentence: “My agent is called [name], built from the [template] template, connected to [tool/data source], and its first job is to [task].”
View transcript
Hello! Your agent is running — now let’s make it good.
Right now, it only knows what’s in its instructions. That’s fine for something simple, but most real work has more nuance — exceptions, formats, reference material you’d hand a new teammate on day one. That’s what skills and a knowledge base are for.
A skill is a repeatable process you teach your agent — steps it follows every time, not just a one-off instruction. A knowledge base is different: it’s where you connect real documents and data, so your agent references facts instead of guessing. It could be a company policy or a reference spreadsheet.
Let’s add both. Think of one part of your task with a clear process, write it out step by step, and add it as a skill. Then upload the document or example you prepared before this lesson as your knowledge base.
Now test it: ask your agent something that requires using that document or following that process. Does it reference the right information? Does it follow the steps in order?
If it gets something wrong, that’s useful — it tells you exactly what to clarify, the same way you’d correct a new hire the second time, not the fifth.
Keep it simple for now. Add what your agent needs for the task in front of it, test it, and build from there.
Next, we’ll get it ready to publish.
Teach Your Agent with Skills
Overview
A working agent is a start — a well-taught agent is what makes it reliable. Skills and a knowledge base are how you pass on the know-how you’d normally explain to a new teammate: the steps you follow, the documents you reference, the exceptions you always make.
In this lesson, you’ll add a skill to your agent and connect it to real information, so it stops guessing and starts working the way you do.
What you’ll learn
- The difference between a one-off instruction and a reusable skill.
- How to add a skill that teaches your agent a repeatable process.
- How to connect a knowledge base so your agent references your real documents and data.
- When to add a skill versus when to just rewrite the instructions.
Before you start
Gather one document, example, or piece of reference material your agent should know about — a policy, a template reply, a spreadsheet, past examples of the output you want.
Completion checkpoint
Write one sentence: “My agent now knows [skill or process], and references [document/knowledge base] to get its facts right.”
View transcript
Hi again! Your agent works, but right now it’s running on the template’s defaults. Let’s customize it to your preferences.
Customizing it is just a conversation. No config screens, no forms. Tell your agent what you want changed, the tone, the format, a rule it should always follow — the same way you’d brief a teammate. It proposes the change, you review it, and you approve before anything is applied.
Try it now. Pick one thing about your agent’s output that isn’t quite right, and tell it what to fix. Look at what it proposes, and approve it or clarify and try again.
Next, decide how your agent runs. You can trigger it manually whenever you need it, put it on a schedule so it runs automatically, or set it to fire when something happens — a new email, a form submission, a record update. Pick whatever matches how the work actually shows up. You can do this by chatting, too.
Once it’s customized and triggered the way you want, it’s time to publish. Publishing makes your agent available beyond just you — to a teammate, your team, or a wider group. Decide who actually needs access first, and start narrow.
Publish, then share it with the people who’ll use it. Tell them what it does, what it doesn’t do yet, and where a human still needs to step in.
That’s it! Congratulations, your agent is customized, triggered, and live.
Customize Your Agent and Publish
Overview
Your agent runs off a template, but it won’t feel like yours until you shape it to your workflow — and you do that just by chatting with it, no forms or setup screens required. This lesson covers how to customize your agent through conversation, set up how and when it runs, and publish it for your team to use.
What you’ll learn
- Customize the agent to your preferences using chat.
- Trigger your agent.
- What “publishing” an agent means, and who can access it.
- Share your agent with your team.
Before you start
Write down one change you’d want to make to how your agent works — tone, format, or a rule it should follow — plus how you’d want it to run: on demand, on a schedule, or triggered by something happening (a new email, a form submission, a record update).