Skip to content

Introduction to Agentic AI and AI Agents · Section 1 · Resources

AI Agents: From Your Idea to Work You Can Use, Complete Resource Pack

Every request, file and link from the lectures in one place. Copy a request with one click.

Arun Nagarathanam · 29 Oct 2026 Guide · 1 of 10 in this course

Before you start

How to use this pack

An exclusive resource for students of Arun's Introduction to Agentic AI and AI Agents course.

This page holds the links, files and requests for each lecture in Section 1. Lecture numbers follow the course outline. The webinar team, Nina, Omar and the manager are a made-up teaching example; the website is Arun's real project. Trying the requests is optional: watching the lectures is enough to follow the rest of the course.

Lecture 1.0

What Could AI Agents Make Possible in Your Work?

See how directing an AI agent can turn a postponed idea, a getting-started page and a welcome email for new customers, into a first version your team can open and judge.

Ideas to keep

  1. 01New customers keep calling to ask how to begin. A clear page and a welcome email could answer them, but making them takes research, writing, design and tools, so the idea stays in a meeting.
  2. 02You can give an agent a goal, the documents that matter and permission to use the files and applications the job needs. It can choose steps, make files, work across tools and bring you a version you can inspect.
  3. 03You respond to that version and the work moves forward. Your team can point to what is missing, decide what to keep and ask for the next version.
  4. 04Once the team has checked the page and email and made them available, a new customer can find the first steps without making the same call.
  5. 05You do not need to code or already know AI terms. The course follows one familiar job first, then shows what lets an agent act, how to direct it and how to recognise a result worth using.
  6. 06You will also see a real website project, aruntastic.com, including how the agent tested the result in a browser.

Try it (optional)

  1. Think of one idea at your work that keeps getting postponed because nobody has time to make the first version.
  2. Name the first thing people could open and react to, such as a page, an email or a short document.
  3. Keep that idea in mind as you watch the webinar example in the next lecture.

Files and links

The real website project you will follow through this section is Arun's own site.

Lecture 1.1

What Can an AI Agent Do Beyond Giving You an Answer?

Follow up a team meeting about a postponed customer webinar, and compare what an answer-only chat gives you with what an agent can save, draft and add for you.

Ideas to keep

  1. 01The situation: your team's customer webinar is next Friday. Nina manages the signup page and Omar prepares the invitation slides, which still advertise Friday and say registration is open.
  2. 02The signup form breaks. Nina can fix and test it by Thursday, but the team wants customers to have several days to register, so it postpones the webinar and will choose a new date once the page works.
  3. 03Your follow-up has three parts: a short written update for the manager (a brief) saved in the shared team folder, a draft of updated invitation slides for Omar to review, and Nina's and Omar's tasks on the online project board.
  4. 04An answer-only chat states the facts correctly, but nothing has been saved, the slides still advertise Friday and the board has no tasks.
  5. 05Given the transcript, the current slides, access to the folder and a working board connection, the agent finds the decisions, marks the date 'to be confirmed', removes the open-registration claim, saves the brief and the slide draft, and adds both tasks.
  6. 06In this course, an AI agent means an application making choices and taking actions toward a goal. The same application may answer a question in one task and do connected work in another.

Try it (optional)

  1. Read the 'What changed?' prompt and the answer the chat gives in the lecture.
  2. List what is still undone after that answer: the brief, the slides and the board tasks.
  3. Read the agent request below and notice that it names the results but not which slide or which part of the transcript to change.
  4. Optional: open the practice files, give an AI application that can read and save files the transcript, the slides and the request, and compare its brief and slide draft with the lecture.
Request · Answer-only chat
What changed?
Use in: An answer-only AI chat, with the meeting transcript pasted in
Request · Prepare the follow-up
Prepare the follow-up from this meeting. Save a short manager brief and a draft of the updated invitation slides in our team folder, and add the agreed tasks to our project board. Leave any undecided details open.
Use in: An AI application that can read and save files in your team folder and is connected to your project board

Files and links

Lecture 1.2

What Does an Agent Need to Work on Your Job?

Work out what each source and each connection lets the agent read or change, and what it still cannot do without them.

Ideas to keep

  1. 01The transcript records why Friday was postponed, Nina's Thursday repair and the decision to choose a date after registration works. The agent needs those facts for an accurate brief.
  2. 02The current slides show the Friday date and the open-registration claim customers would see. Those are the parts the agent changes in Omar's draft.
  3. 03You choose the transcript and slides, because you know which meeting and which customer event the job is about. Without the transcript the update can be polished but miss the decisions. Without the slides the invitation stays unchanged.
  4. 04Reading a file and changing it are different. Saving the brief and the slide draft needs permission to change files in the team folder.
  5. 05The project board needs its own working connection and permission to add tasks. Without it, the agent can still write task descriptions for you to add. Check the board to see which result you actually got.
  6. 06No connection can supply a decision the team has not made, so the new date stays marked as undecided.
  7. 07A draft email to registered customers can be reviewed now. Sending it is a separate action that waits for your team's approval.
  8. 08On the website project, the agent used the site's page files and notes and changed a test copy Arun could open in a browser, while visitors kept using the public site.

Try it (optional)

  1. For each item, say whether it gives the agent information or a place to make a change: the transcript, the slides, the team folder, the project board.
  2. Imagine the board connection is off. Say what the agent can still produce.
  3. Decide which action in this job should wait for your team's approval.

Files and links

No outside links in this lecture. The transcript and slides the agent reads are in the webinar practice files.

Lecture 1.3

Can You Guide an Agent Before You Know the Final Result?

See how a rough request gives you a version to judge, and how one round of specific feedback makes it useful to a busy manager.

Ideas to keep

  1. 01A rough request can produce a brief where every fact is correct but the order follows the meeting, so the manager reads the page problem and the repair before finding out the webinar is postponed.
  2. 02If your manager only reads the first two sentences, those sentences should hold the decision and who is responsible.
  3. 03Specific feedback names what goes first: the postponed webinar and the undecided date, then Nina's repair and Omar's slides, then the reason.
  4. 04The same facts become more useful because you changed their order.
  5. 05Clear direction helps from the start. Rough direction still gives you a version to judge, followed by specific feedback.
  6. 06On the website, early visual directions looked dated to Arun. He gave the agent website-design videos showing styles he preferred, chose and refined a later direction, approved a test copy and then made a separate decision to make the site public.

Try it (optional)

  1. Read the rough request and the first brief it produced in the lecture.
  2. Before you read the feedback below, decide what you would put in the first two sentences.
  3. Compare your choice with the feedback and with the revised brief.
Request · A rough first request
Turn the planning meeting into a short update for the manager.
Use in: An AI application that can read the meeting transcript
Request · Feedback that reorders the brief
Lead with the postponed webinar and the undecided date. Then name Nina's Thursday page repair and Omar's review and final revisions to the invitation slides. Put the registration problem and reason for postponing after those decisions.
Use in: The same conversation, after the first brief

Lecture 1.4

How Can an Agent Check the Work It Has Done?

Ask an agent to check its own drafts against the source, then watch an agent test a real web page like a visitor and spot-check what it reports.

Ideas to keep

  1. 01Before handing work over, ask the agent to compare the brief and slide draft with the transcript and report any mismatch, missing decision or undecided detail shown as confirmed, with the source passages.
  2. 02If a draft announces a new webinar date, the transcript shows the team has not chosen one. You get a specific correction to review instead of hunting for every difference yourself.
  3. 03Read the report, then spot-check a fact that matters, such as Nina's Thursday deadline. The agent helps with the comparison; you decide whether the result is ready.
  4. 04A website needs checks of what a visitor can do, not only of the words. With a browser, the agent can choose which links to follow and which buttons to try, so you do not have to describe every click.
  5. 05In the recording, the agent tests the student resource library for Arun's SEO course. Choosing Toolkit leaves only toolkit resources. The agent also opens the main toolkit on its own.
  6. 06The agent flags 'Fix the Technical SEO Problems That Google Hates': its description calls it a toolkit but it is tagged Guide, so the Toolkit filter hides it.
  7. 07If a test finds a problem, ask the agent to fix it and repeat the same check, such as correcting a claim and comparing it with the transcript again, or repairing a button and trying it again.

Try it (optional)

  1. Read the comparison request below and name the three kinds of problem it asks the agent to report.
  2. Open the SEO course resource library from the website page and choose Toolkit, as the agent did.
  3. Look at what remains and open the main toolkit, 'Course Tools & Resources: Your Complete Reference Guide'.
  4. Find 'Fix the Technical SEO Problems That Google Hates', the resource the agent flagged, and check its tag and description yourself.
  5. Optional: give the visitor-test request to an AI application with a browser and point it at a website you look after.
Request · Check the drafts against the transcript
Compare the manager brief and updated invitation slides with the original meeting transcript. Report any mismatch, missing decision or undecided detail presented as confirmed. Show the source passages behind your findings.
Use in: An AI application that can open the brief, the slide draft and the transcript
Request · Test a website like a visitor
Test my website like a visitor. Go through the main pages, follow links and try the buttons. Tell me what you checked, what worked and what needs fixing.
Use in: An AI application with a browser, such as Claude Code in the Claude desktop app

Files and links

Lecture 1.5

What Could You Put an Agent to Work On?

Break a larger customer-help project into jobs you can name and direct, and see how the same approach repeats for the next webinar meeting.

Ideas to keep

  1. 01Start with what you already have: your company's existing instructions and the questions customers keep asking. Ask for a one-page guide to the first steps, saved in the shared folder.
  2. 02Judge the guide with plain questions. Does it answer those customer questions? Does it put the first action first? If it leaves out how to get help, point to the support information and ask the agent to include it.
  3. 03Ask the agent to compare the revised guide with the company instructions, then check the answers customers are most likely to rely on.
  4. 04Once you are happy with the guide, ask the agent to prepare the website page and a welcome email that tells new customers where to find that page. Your team reviews both before anything is published or sent.
  5. 05The larger project is four jobs you can name: write the guide, check its instructions, make the page and prepare the email.
  6. 06On the website redesign, the agent changed the home page, course pages and articles and helped create and place 211 refreshed article images. Arun compared designs, chose how to present his courses and tools, asked for a visitor test and approved making the site public.
  7. 07A familiar job repeats: after the next webinar planning meeting, give the agent the new transcript and current slides. If a date has now been chosen, the update and slides should show it.

Try it (optional)

  1. Pick one larger job at your work that people keep postponing.
  2. Write down the material you would give the agent, such as instructions, common questions or a meeting record.
  3. Split the job into named pieces you could judge one at a time.
  4. Mark the step where your team reviews before anything is published or sent.

Files and links

More in this section

Keep going

Keep building

More from Aruntastic

Practical courses for people who want to build useful things with AI.

Toolkits and programme

Go deeper

Claude certification practice

Preparing for a Claude certification?

Practice exams written by Arun: 600 questions across 4 mock exams and 2 drill banks. Unofficial preparation, not affiliated with Anthropic.

Keep going

This resource is part of an 11-section course.

Follow the lectures on Udemy, then come back here whenever you want the files, requests and notes for a section.

Open the course Introduction to Agentic AI and AI Agents · Udemy