October 14, 2026 • Gurus AI — Agents at Work What are you and your AI agents working on today? I’m Don Wansley. Call me Guru. Today’s example is for anyone organizing a community gathering, class, business meetup or family event. ❓ Example question: “I have event notes on my phone. Can an agent help me invite people?” Yes. Give it your confirmed details and ask it to turn those notes into a clear invitation. Here’s a practice example—not an announced Gurus AI event. 📝 Example notes Event: Neighborhood Business Meet & Greet When: Saturday, 10 AM–noon Where: Community Center, Room 2 Who: Local business owners Bring: One task you want help completing RSVP: Reply YES 📱 Copy this assignment “Turn these event notes into a friendly text invitation under 80 words. Include the event, time, place, who it is for, what to bring and RSVP instruction. Use only my notes. Then give me a shorter reminder version. Flag missing details. Show both drafts for my review.” ✅ Example result “You’re invited to our Neighborhood Business Meet & Greet this Saturday, 10 AM–noon, at the Community Center, Room 2. Meet local business owners and bring one task you want help completing. Want to join us? Reply YES!” Short reminder: “Saturday: Neighborhood Business Meet & Greet, 10 AM–noon, Community Center, Room 2. Bring one task you need help with. Reply YES to RSVP.” 🔎 How you use it Check the date, address and RSVP details against your real event notes. Replace the practice information with your confirmed information. Copy the invitation into your chosen message app and send it to people who welcome it. If you want an agent to send it, it needs connected tools and your clear instructions about recipients and sending. 🤝 Your turn What is one task you need help completing? • Already using an agent? Tell us its job and what it accomplished. • Getting started? Describe the task you want to hand over. • Stuck? Tell us where you need help. • What would you like to build next? Try this with your own notes today. Share the invitation you checked, with private details removed. That finished result is your Proof of Work.

October 13, 2026 • Agent Out the Box practical lesson Yesterday we turned one safe run into a reusable job card. Today, let’s check whether that card works when the inputs change. I’m Don Wansley. Call me Guru. 🎯 Your result today One tested job card, one checked draft, and a short Proof of Work receipt. You can do this from Android, iPhone, or a computer. 🧪 Give the same job three small tests Use invented practice details and the same instructions for each test. Example job: prepare a customer FAQ from approved service notes. 1️⃣ Complete inputs Provide a service name, description and confirmed price. Check that the draft preserves those facts and follows your requested format. 2️⃣ Missing information Leave the price out. Check that the agent marks it “Needs confirmation” instead of inventing a price. 3️⃣ Conflicting information Provide two different prices for the same service. Check that it identifies the conflict and asks which source is current before completing that answer. 🛠️ Copy this assignment “Use this job card and the practice notes to prepare a customer FAQ draft. Use only the supplied facts. Mark missing information as ‘Needs confirmation.’ If sources conflict, identify the exact conflict and pause the affected answer. Return the draft, the source for each answer, and a checklist showing which job-card requirements passed or failed. This practice authorizes draft preparation only.” 🔎 Inspect the work yourself Compare the draft with the notes. Did the confirmed facts stay accurate? Did missing facts remain visible? Did the conflict produce a precise question? Did the agent stay inside its boundaries? An agent’s own checklist helps you review; your comparison supplies the evidence. 🔁 Correct, rerun, remember If a check fails, update the specific instruction that caused the problem. Rerun the affected test. Save the revised card and the test receipt together, then provide them when the next session begins. A successful practice run supports the next small assignment; broader authority still requires your explicit authorization.

Yesterday we gave our Digital Employee one small task, a clear boundary and a result we could check. Today, let’s turn that successful run into a job it can repeat. I’m Don Wansley. Call me Guru. 🎯 A good result is not a workflow yet One successful run proves that an approach can work once. A repeatable job needs the useful parts captured clearly enough to use again. Do not save only the final answer. Save the instructions, approved inputs, boundaries and checks that produced it. That becomes your job card. 🗂️ Build a six-part job card Give the job a simple name, then record: 📌 Result — What should exist when the job is finished? 🔔 Trigger — What tells the Digital Employee to begin? 📎 Inputs — Which approved notes, files or links may it use? 🛠️ Actions — What steps should it follow? 🚧 Boundaries — What must it pause before sending, publishing, purchasing, deleting or changing? ✅ Proof — What evidence shows that the result is correct and complete? Example job: Prepare a customer FAQ draft. Trigger: Guru provides approved service notes and requests the FAQ. Proof: Every answer traces back to the notes, missing facts are marked “Needs confirmation,” and nothing is sent or published. 🧠 Improve the card with real experience Use the receipt from the first run. Ask: • Which instruction prevented a mistake? • Which input was missing or unclear? • Which check caught a problem? • What should stay the same next time? Add the answer to the job card. Keep the instructions short enough to follow and specific enough to test. Try this assignment: “Review the completed work receipt and turn it into a reusable job card. Include the result, trigger, approved inputs, actions, boundaries and Proof of Work. Preserve the rules that worked. Mark any missing decision as ‘Needs Guru’s confirmation.’ Prepare the card for review; do not run the job or take an external action.” 🧪 Test before you scale Run the job card on one new, low-risk example. Compare the new result with the first one.

Yesterday we checked whether our Digital Employee had the assignment, inputs, access, destination, authority and finish line it needed. Today, let’s use that readiness check to run one small job safely. I’m Don Wansley. Call me Guru. 🎯 One task. One boundary. One result. A first run should be small enough to inspect from beginning to end. Do not start with “handle my whole business.” Choose one useful job with a result you can check. Examples: ✅ Turn approved notes into a short FAQ draft. ✅ Organize one lesson into clear sections. ✅ Summarize one meeting transcript and list the open decisions. ✅ Prepare one social post for review without publishing it. The goal is not to prove that the agent can do everything. The goal is to collect evidence about how it handles one responsibility. 🧭 Set the safe working lane Give your Digital Employee four things before it starts: 📌 Task — What single result should it create? 📎 Approved inputs — Which notes, files or links may it use? 🚧 Boundary — What must it not send, publish, purchase, delete or change? ✅ Check — What will you inspect before accepting the result? Try this assignment: “Use only the approved notes I provide to prepare one customer FAQ draft. Do not add prices, promises or policies that are not in the notes. Mark missing information as ‘Needs confirmation.’ Do not send or publish anything. Return the draft, list the source used for each answer, and tell me what remains uncertain.” 🔎 Watch the first run like a coach Do not judge the result only by whether it sounds good. Check whether the Digital Employee followed the job: ✅ Did it use the approved inputs? ✅ Did it stay inside the boundary? ✅ Did it mark uncertainty instead of guessing? ✅ Did it deliver the requested format? ✅ Can you trace the result back to the source? If one check fails, correct the instruction and run the small test again. That correction becomes part of the reusable job card. 🧾 Save the first-run receipt Record: • Job attempted: ____

Yesterday we gave our Digital Employee clear reasons to stop and ask. Today, let’s check what it needs before the job begins. I’m Don Wansley. Call me Guru. 🎯 Scheduled → Ready → Executed → Verified Each stage needs its own evidence. Putting a job on a calendar sets the time. A readiness check confirms that the agent has the instructions, inputs, access and destination it needs. Execution does the work. Verification checks the result. 🧠 Give your Digital Employee a coach’s checklist Before a job starts, ask it to confirm: 📌 Assignment — What exact result should exist? 📎 Inputs — Can it actually open the approved notes, image, video or file? 🛠️ Access — Is the correct account and tool available? 📍 Destination — Where should the result go? 🔐 Authority — Which actions have you authorized? ✅ Finish line — How will it prove the job worked? The attached AI Coach image illustrates step-by-step guidance. Its example commands are not instructions to run on your device. Use a checklist that fits your actual job. 📱 Try one job from your phone Choose a small task such as preparing a lesson summary from your own approved notes. Give your Digital Employee this assignment: “Before starting, check that you can read my attached notes and understand the intended audience. Prepare one short lesson summary with a practical action. List missing inputs or access. Use only supported information. Show the finished draft and explain how you checked it. This assignment authorizes draft preparation only.” If you want a later external action, give it clear authorization for that action and destination. 🚧 Recover without creating duplicates If an authorized job fails partway through, first check what already exists. Continue from the unfinished step. An uploaded image with an unpublished post needs a different next action from a post that is already live. Ask for a short receipt: • Completed: ____ • Checked: ____ • Blocked: ____ • Next action: ____ 🤝 Learn from real work Gurus AI grows through my ongoing collaboration with ChatGPT and our use of OpenAI technology. Preparing, executing and checking real assignments gives us experience to teach and a work record that can support earned trust.