Transforming teaching, learning and leadership through the strategic application of technology has been Miguel Guhlin’s motto. Learn more about his work online at blog.tcea.org, and mguhlin.org. Catch him on Mastodon at @mguhlin@zirk.us
Areas of interest flow from his experiences as a district technology administrator, regional education specialist, and classroom educator in bilingual/ESL situations. Learn more about his credentials online at mguhlin.org.
Are you looking for something lighter than a rubric or pre/post test, but still useful in PD sessions? Perhaps you want to vibe code your way to a standalone icebreaker personality quiz you can use with your students? That’s how I started. I wanted a fun, low-stakes icebreaker, a way for educators to reflect on their teaching style. Of course, I also wanted to test out another vibe coding solution, using Gen AI to develop a website, interactive personality quiz. In this two part blog entry, I’ll share how to build a custom GPT and then how to turn it into a standalone website. You might start with a personality quiz, then graduate to an online entry/exit ticket or icebreaker.
In this part, we’ll explore first how to create a custom GPT. Then, in part two, we’ll turn it into an interactive webpage you can host for free on GitHub. Finally, in part three, I’ll share how you can save the data from the interactive webpage to Google Sheets.
Creating an Engaging, Fun Personality Quiz
One of my favorite personality quizzes online is, “Which character are you?” You have probably taken a few of these on Facebook. What bothers me about those quizzes is the collection of personal data. A question in my mind was, “How could I design a privacy-safe, fun personality quiz for teachers?”
Winnie the Pooh seemed the best place to start. I love the characters of the 100-acre wood, and my family often identifies with Pooh and characters. Are you a Tigger or an Eeyore? Maybe you are Owl or Roo?
By popular demand from my work colleagues, I also wanted the quiz to sort educators into characters they were familiar with. So, in addition to Winnie the Pooh, they wanted:
The Smurfs
Snow White and the Seven Dwarfs
Looney Tunes
Each quiz would end with a playful answer: “You’re Tigger as a teacher,” or “You’re Brainy Smurf in the classroom.” The results included a short description of what that character is like as a teacher.
Starting with the Problem, Not the Tool
While I wish I could say I began with the end in mind, I had only a vague idea. I started a conversation with ChatGPT and the adjustments came over time. After working with colleagues, I had to add different universes (e.g. Smurfs, Looney Tunes) because I had different colleagues ask, “Could you please make an assessment for Looney Tunes or Smurfs, etc.?” The first one I started with included Winnie the Pooh characters. At the end, I ended up with several additional considerations:
Each quiz uses a single universe (Pooh, Smurfs, Snow White, or Looney Tunes).
Each quiz has exactly five multiple-choice questions.
Questions appear one at a time.
The quiz waits for the user’s answer before moving on.
At the end, the GPT assigns one character based on the most frequent trait pattern.
The result explains what kind of teacher that character would be.
Again, these developed over time via an interactive conversation with ChatGPT. Then, I asked ChatGPT to give me the custom instructions for the GPT. At that point, I had something to work with.
Want to try this out on your own? If you have a paid account for an AI chatbot (such as BoodleBox Unlimited, ChatGPT Plus/Teams/Education/Enterprise, Google Gemini Gem, Claude Project/Artifact), you can use the custom instructions in this Google Doc to get started. Simply save the instructions as a text file or markdown content with a filename extension of TXT. The filename and extension would be instructions.txt.
Designing the Quiz Flow with ChatGPT
Next, I focused on behavior, not characters. I asked ChatGPT to sketch the flow of a single quiz from start to finish. The prompt looked something like this:
Help me design a personality quiz flow for educators.
Five questions, multiple choice.
Ask questions one at a time.
Wait for an answer before showing the next question.
At the end, analyze answers and assign a single character type.
Make it easy to convert into Custom GPT instructions later.
ChatGPT responded with a simple sequence:
Greet the user and explain the quiz theme.
Ask Question 1 with answer options A–D.
Wait for an answer.
Record the trait associated with that option.
Repeat for Questions 2–5.
Tally which trait pattern appears most.
Map that pattern to a character.
Present the result with a short, teacher-focused description.
This became the framework for every quiz, no matter the universe.
Building the Character Library
As I mentioned earlier, some of my colleagues wanted to know which Smurf or Looney Tunes character they would be. The chatbot provided the descriptions of each character. Of course, the magic is that this would happen no matter what universe they are in:
Winnie the Pooh: Kind, gentle, loyal. As a teacher: Warm, welcoming, always encouraging.
Piglet: Timid, anxious, brave when it counts. As a teacher: Patient, supportive, nurturing confidence.
Tigger: Energetic, optimistic, adventurous. As a teacher: Fun, engaging, brings excitement to learning.
Rabbit: Organized, responsible, a bit fussy. As a teacher: Structured, caring, focused on progress.
I repeated the process for:
The Smurfs (Papa Smurf, Smurfette, Brainy, Hefty, etc.)
Snow White and the Seven Dwarfs (Snow White, Doc, Grumpy, Happy, etc.)
Looney Tunes (Bugs, Daffy, Road Runner, Wile E. Coyote, etc.)
Each question’s options pointed to the traits of one or more characters. At the end of five questions, the most frequently tapped trait pattern determined the result. Describing this takes longer than the actual amount of time spent asking the Gen AI chatbot to do this.
Turning the Model into Custom GPT Instructions
Once the Project worked well, I asked it to give me the custom instructions (a.k.a. system prompt) for the Custom GPT.
Take our quiz flow and character descriptions and turn them into Custom GPT instructions.
The GPT should offer a menu of quiz themes (Pooh, Smurfs, Snow White, Looney Tunes).
Once the user chooses one, it should explain the quiz and start with Question 1.
It must not skip ahead. Wait for each answer.
After five questions, it should assign a character and explain what kind of teacher that character would be.
Always keep the tone warm and educator-friendly.
Display this image at the start:.
Again, with some interactive back-and-forth with the chatbot, ChatGPT generated a set of “rules.” Those rules included:
How to greet users
How to present the image
How to guide quiz selection
How to enforce the one-question-at-a-time rule
How to tally responses and deliver a result
This became the backbone of my Custom GPT, ensuring a reliable response each time. You can see the custom instructions here that I relied on.
Testing, Tuning, and Making It Teacher-Friendly
The last step was trial and error. I found volunteers to run through the various iterations and it turned out to be a wild success. People love fun personality quizzes. As I tested it, I saw different ways to improve it, such as:
Simplifying question wording
Balancing answer choices
Softening language for more hesitant audiences
Ensuring the final descriptions felt affirming, not judgmental
At the end, I had a Custom GPT that educators could enjoy in a PD session, department meeting, or just for fun. The quiz is not about labels. It is about language for talking about how we show up in the classroom. It also served as a way to introduce people unfamiliar with Gen AI to a custom GPT.
Pondering Next Steps, an Interactive Webpage
The next step became obvious after I tried it with others who lacked a ChatGPT account. How could I move this beyond the OpenAI Custom GPT space? Doing so would allow anyone to take the personality quiz via a simple link, no ChatGPT account required. What’s more, I’d be able to create similar assessments that could avoid OpenAI’s strictures of adult learners only. In part two of AI Maker Magic, you will see how to use vibe coding to turn this quiz into an interactive webpage.
Every day, a new Gen AI tool (see my working list) lands in my inbox or social media feed. The constant stream of new technologies can be overwhelming. If you’re feeling Gen AI fatigue, you’re not alone. As educators, a deluge of AI tools drops on our heads daily. Each claims to save time. Each is guaranteed to improve student achievement. Without a structured approach, all of these tools can become time traps rather than time savers.
Mark Zimmerman wisely cautions in his CIO piece, How AI Tools Can Become a Time and Cost Trap, that we must set clear boundaries. Otherwise, we risk spending countless hours experimenting with AI. That’s a problem given our limited time and funds. Remember, as educators, we have to implement new tools in alignment with research-aligned ways. What’s more, how do you know which are FERPA-compliant and offer data privacy agreements appropriate for K-12 education? Let’s explore five strategies that can help you stay in control.
Strategy #1: Setting Clear Goals with Success Criteria
Just as we teach students to set clear learning targets, educators need specific goals when implementing AI tools. Before diving into any AI tool, ask yourself: “What specific educational problem am I trying to solve?” This simple question, can help students focus on what’s productive, and what’s not.
Consider creating an AI Goal Sheet before each session that includes:
The specific problem you’re addressing
Time limits
Clear success criteria for evaluating outcomes
One middle school science teacher uses a simple log to track her AI experiments: tool used, time spent, and specific benefits achieved. Use of a tool like this can eliminate “going down the rabbit hole” and adds a sense of urgency.
Strategy #2: Feedback-Informed Tool Selection and Mastery
Rather than constantly chasing the newest AI tools, focus on several Gen AI tools that deliver results. Remember the Greek poet Archilochus’ adage of the fox and the hedgehog. That is, “the fox knows many things, but the hedgehog knows one big thing.” Or as Paul Roetzer and Mike Kaput, hosts of The Artificial Intelligence Show podcast, assert, “Learn one Gen AI tool really well” (a paraphrase).
Get to be an expert on one tool, and solicit feedback from other educators as to which works best. Get comfortable with that one Gen AI tool before moving to another. What does expert mean? Ask yourself, “How does this AI tool work in my situation?” Explore (keep strategy #1 in mind) until you identify a use case. A use case is a specific process that can be translated into Gen AI tool to realize some improvement.
Consider adopting these implementation approaches:
Select one AI tool per quarter to learn deeply, accepting being uncomfortable is part of that process
Create feedback forms for students: “How did this AI-enhanced lesson help your learning?”
“I was spending hours every week testing different AI tools,” admits a high school English teacher I chat with from time to time. “Now I focus on getting to know one tool that has become my go-to. I’m spending time going deep in applying it to my work, and have reclaimed my planning time as a result.”
Strategy #3: Metacognitive Integration Planning
How are you using Gen AI in your work? Take the time for structured reflection. It’s hard to do because it’s easy to keep moving forward to new or different projects. But take a quiet moment to jot down a few reflections. Differentiate between problem-solving and mere curiosity-driven experimentation.
Try these practical applications:
Weekly reflection: “Did my AI use this week directly improve student learning outcomes?”
Monthly team meetings to share AI successes and challenges
School-wide AI integration guidelines with clear evaluation criteria
One technology director suggested this approach:
“Instead of having every teacher experimenting independently, we now have a structured approach where we collectively evaluate tools against our learning objectives. The result is a more strategic implementation and less wasted time.”
Strategy #4: Collaborative Learning Through AI Peer Teaching
“Would you mind sharing what you did, Miguel?” That’s the question I loved to hear from a grade level colleague in my small East Texas school. As a third-grade bilingual teacher working in the portable building, I loved being able to share what my students created. And, of course, that involved me sharing my lesson activity with technology enhancement.
Make sharing your Gen AI a systemic action at the grade level, campus, and district level. This can go a long way to developing a centralized AI strategy. Not only do you learn what others really think about Gen AI tools, all get to provide solutions that work. That’s more important than AI hype or hate on social media. One thing to keep in mind? You won’t need to do this forever.
Consider implementing:
Identify AI Champions in each grade level or department who research and share effective tools
Monthly “AI Show and Tell” sessions where educators demonstrate successful applications
Shared repositories of proven AI prompts and strategies for common educational tasks
What I love? The idea that peer teaching is the best source of learning about new technologies. Well, maybe after TCEA professional learning solutions.
Did You Know? Get access to BoodleBox Unlimited (pro version) for six months while you complete the Educator Accelerator certification course, The AI-Amplified Educator. Sign up now.
“Our AI Champions program has transformed how we approach technology,” reports an elementary principal. “Instead of everyone struggling independently, we now have a collaborative system where expertise is shared efficiently.”
Strategy #5: Balanced Skill Development with Transfer Strategies
Can you still maintain and develop traditional teaching skills? Yes, but also consider strategically integrating AI. Don’t rely on technology alone. Reach for paper and pen, too.
Practical applications include:
Alternate between AI-assisted and traditional lesson planning
Maintain a balance of tech-enhanced and unplugged learning activities
Routine engagement in non-AI professional development to sharpen core teaching skills
A veteran teacher shared her balanced approach: “I use AI to help generate differentiated practice problems. I still design the core learning experiences myself. This balance ensures I maintain my instructional expertise while leveraging AI for efficiency where it makes sense.”
Recommended Education-Focused AI Tools
To implement these strategies effectively, consider exploring these education-focused AI platforms:
BoodleBox for collaborative AI exploration with built-in educational safeguards
MagicSchool for streamlining lesson planning and assessment creation
SchoolAI for creating personalized learning experiences
It’s important, Zimmerman says, to measure the actual benefits. That’s why keeping a log of time spent with specific results can be helpful. I like to jot these down in a paper notebook to ensure long-term memory retention occurs. In fact, I have a dedicated notebook (perhaps, two or three notebooks now) with my notes.
Success Indicator
Measurement Method
Example
Time savings
Before/after time tracking
Lesson planning reduced from 2 hours to 45 minutes
Effective AI integration isn’t about using the most tools or the newest technologies. When you find and use tools that enhance learning, you become more effective. Avoid AI time traps. Use one of these five strategies to get started. You will find your confidence improving.
What stops students from delving deep into learning, mining for more glittering mental riches? Or, taking the time to craft a well-written piece? It doesn’t matter now. Generative AI has short-circuited thinking and writing as assessment. In this blog entry, I’ll share an approach to supporting student self-efficacy that caught my eye, “Viva Voce.”
“The cheating is off the charts. It’s the worst I’ve seen in my entire career, “ said Valencia High School English teacher Casey Cuny, a 23-year veteran. “Anything you send home, you have to assume is being AI’ed,” he said.” (Source: Larry Cuban on School Reform and Classroom Practice, 10/6/2025)
Supporting Student Self-Efficacy
In their book, The Opposite of Cheating, Tricia Bertram Gallant and David A. Rettinger offer several reasons why students rely on Gen AI. The authors offered ten principles.
The most important one? The suggestion that knowledge is a learner’s construct. Learning is a process of constructing knowledge, assembling something that connects to a student’s schema.
The real enemies of student constructed learning that they own? Those enemies include stress, limited time, and fear of failure. Students embrace Gen AI tools to quicken their work since they lack self-efficacy. The more I reflected on their arguments, the more they made sense. But students are embracing Gen AI, as much as teachers are. AI makes traditional assessment easy on both sides.
What May Work Instead
If convenient assessment forms have quit the field, then let us think different. But where can beleaguered educators find assessments that work in spite of AI?
Handwriting facilitates long-term information retention
Handwriting makes it easier for me to give voice to ideas recorded, one word at a time
Handwriting connects new learning to my schema, making it easier to recall and share
We know handwriting notes works, but so does giving voice to ideas.
A Classroom of “Living Voice”
Viva voce comes from Latin. It means, “with living voice.” The term refers to oral examinations where students speak their explanation rather than type it.
Students need oral assessment skills. They need that skill for job interviews, seminars, and professional presentations. It is a skill that I wish I had developed early on rather than tacked on after high school.
How can we teach viva voce in K-12? Let me suggest VIVA, a framework for viva voce in K-12.
A Framework for Viva Voce Success
VIVA offers four steps that work across grade levels and subjects. Let’s take a closer look.
V – Verify. Students begin with a clear definition or statement of the concept.
I – Illustrate. Students provide concrete examples, scenarios, or demonstrations. The goal is to connect abstract concepts to real applications. This is about making their ideas, their thinking, visible to all.
V – Validate. Students share their reasoning, providing justifications for their understanding. This involves presenting evidence, logic, or supporting arguments. The goal is not to memorize, but to comprehend and share from that grasp of content.
A – Apply. In the final step, student apply their knowledge in new contexts or solve new problems. If you can think with a concept, applying it to a new situation, you show transfer learning has occurred.
Key concepts in PRISM and the SOLO Taxonomy frameworks are well-represented. What’s more, making an anchor chart of VIVA is easy.
Post the four steps where it can students can see it at all time with language that students understand:
– V: Say what it is
– I: Give an example
– V: Tell why it works
– A: Use it in a new way
How can you prepare students to use VIVA? The anchor chart and modeling are two strategies. Here are some more.
VIVA Preparation Strategies
These strategies make using VIVA easier for students. The strategies include:
Transition phrases support transitioning to the Illustrate, Validate, and Apply steps of VIVA. This reminds me of an excellent text for teaching writing, They Say/I Say(listen to podcast). Use strategies like this to build comfort and skill with the framework.
Why It Matters
Embracing VIVA has a host of benefits, such as developing communication skills. One of the challenges I face when speaking involves real-time organization of thoughts. In time, verbal confidence grows…you know that you can organize on your feet. Teachers can probe deeper understanding through follow-up questions and spontaneous explanations. Metacognition and alternate ways of demonstrating knowledge are critical, too.
Let’s take a look at an example below.
Transform Show-and-Tell with Structure Young students already take part in show-and-tell. Use VIVA to prepare students to think on their feet. Here’s a scenario:
Ms. Cruzes’ second grade students bring an object from home. Instead of free-form sharing, they use the VIVA oral framework.
Verify: “This is my grandmother’s compass.”
Illustrate: “It has a needle that always points north. See how it moves when I turn?”
Validate: “Compasses help people find directions because the needle is magnetic.”
Apply: “If I got lost in the woods, I could use this to walk north toward the road.”
Here is one K-5 example of VIVA. As you read them, ask yourself, how might you do something similar in your classroom?
Make Oral Assessment Work, Prepare Your Students
It’s nice to imagine that someday AI Viva may be a reality. Coaching students in VIVA oral framework use pays off. Whether AI takes over K-12 or not. Oh, I almost forgot. Here are two Gen AI assistants you can use to run any topic through to get a VIVA-formatted example.
Video-based coaching, now known as micro-teaching, isn’t just another ed tech trend. According to John Hattie’s research, it has an effect size of 0.99 on teacher growth. This suggests it has the potential to accelerate or even double the rate of professional learning. Despite this evidence, however, many may still view the camera as a tool of judgment rather than transformation.
Micro-teaching: A technique in which a teacher delivers a short, recorded lesson that is then analyzed by the teacher and/or other teachers and leaders for the purposes of improvement. The lessons are usually videotaped to assist with the subsequent analysis. (source)
“I don’t want to see myself on camera. I already know I talk too much,” said one teacher. What changed her mind? Making the shift from focus on her to finding evidence of student growth. We can all overcome our shyness and fear in front of the camera when we focus on student learning.
From “I Feel” to “I See”
A significant challenge in instructional coaching is the gap between what a teacher remembers happening and what actually happened. Our brains naturally fill in gaps, creating a subjective narrative that often differs from reality. A teacher might say, “I feel like the lesson went well,” but video reveals a different story: 40% of students were off-task during transitions and/or the teacher called on the same five students throughout the period.
Video changes the conversation. By focusing on objective evidence (student talk ratios, engagement patterns, pacing, body language), we move from feelings to facts. This isn’t about catching mistakes. It’s about capturing excellence and making the invisible visible.
The EIIR Framework: Your Roadmap for Growth
To make video coaching sustainable and effective, you might consider using the TCEA EIIR Framework, a continuous cycle of improvement:
1. EMPOWER: Set clear, meaningful goals together using the RISE framework (Relevant, Inspiring, Specific, Evidence-based). The teacher owns the process from the start.
2. IMPLEMENT: Record practice. Here’s the key: short clips of 10-15 minutes work best. Full-hour recordings are daunting and difficult to analyze. Start small.
3. INVESTIGATE: Analyze the footage using structured protocols like PRISM (Purpose, Reflection, Insight, Strategies, Monitoring). This keeps the debrief focused on instructional impact rather than personal appearance.
4. REALIZE: Celebrate the “Glows,” identify the “Grows,” and set new targets. This cycle repeats, creating continuous improvement over time.
Of course, you can always use your favorite coaching cycle, such as Diane Sweeney’s Results-Based Coaching Tool or Jim Knight’s The Impact Cycle. Or an eclectic combination that your organization decides is better.
The PRISM Protocol: Keeping Feedback Focused
You may be familiar with PRISM, which often focuses on multiple levels of the SOLO Taxonomy. PRISM often starts out with completely different definitions. However, when debriefing video, I thought it might be fun to adapt it as a way to ensure conversations stay productive:
PRISM Element
Core Question
Coaching Focus
Patterns
What patterns do you notice? What general patterns help us understand what’s going on in the classroom?
Teacher-led observation before coach input
Reasoning
How do things fit together? What makes sense when you look at everything? What was the learning goal?
Alignment between objective and instruction
Ideas
What surprised you? What other ideas should we explore? What happens when we try new ways of thinking?
Moments of discovery about student learning
Situation
How does what happened connect to instructional strategies?
Evidence-based instructional moves
Methods
How will you track progress? How can we check progress? How do we know which answer is right?
Next steps and success indicators
This structure moves us away from subjective feelings (“I think I did okay”) toward objective evidence (“I see that only three students responded during the discussion”).
Overcoming the “Cringe” Factor
“I’m glad to see you started with the fear factor about video in the classroom,” says Brian Lamb of SWIVL. We were reflecting on the fact that watching yourself on video is uncomfortable. You know, I’m right there with anyone who worries about how they look, how they sound, whether their hair is messy (well, I don’t have that problem now!). These concerns are valid, but they’re also distractions from what matters most. That is, student learning.
One way to bridge the gap between distractions and our focus? Build a Trust Bridge. You can do this by taking the following actions:
Model Vulnerability: Film yourself first and share it with your team. I started with audio, then slowly moved up to video to capture my body language, too. Those non-verbal cues really can send their own message or amplify your spoken words. When I did this, I discovered I said “um” 47 times in a 15-minute lesson. Sharing that vulnerability opened the door for others.
Focus on Students: Pivot the conversation away from the teacher’s appearance and toward student reactions. Ask: “What do you notice about student engagement during the transition?” instead of “How did you feel about your explanation?”
Ensure Safety: Establish a “We Will” commitment. The teacher owns the footage. It’s used for growth, not evaluation. Confidentiality is non-negotiable.
Start Small: Use the “Micro-Coaching Sprint” approach. Record just 10-15 minutes. Choose a low-stakes moment. Build confidence before tackling more complex lessons.
A quick note: We’ve all worked in low-trust environments. At some point, you make a decision to separate your teaching actions from your personal ego. You can always get better. As a professional, you trust yourself to always be improving and deal with critiques that may or may not be accurate. Let the chips fall where they may.
Practical Tools: From Smartphone to Swivl
Getting started with video-based coaching doesn’t need to cost an arm and a leg. You don’t need a Hollywood budget to start video coaching. Here’s your Mobile Prep Sprint checklist:
[ ] Storage: Ensure 2GB of free space on your device
[ ] Battery: Charge to at least 80%
[ ] Do Not Disturb: Activate to prevent interruptions
[ ] Positioning: Place camera where it captures both teacher and students
[ ] Audio Check: Test sound quality in the actual classroom
For those ready to level up, tools like the Swivl M2 robot can automatically track a teacher’s movement, ensuring the “pilot” of the lesson stays in frame.
Creating Questions from Video: AI-Powered Analysis
Once you’ve captured video, you can extend its impact by creating reflection questions or student assessments. I like to transcribe audio and/or video, then run it through an AI chatbot for analysis. While tools like SWIVL offer various frameworks, you can also use some like the TCEA EIIR framework for that. Explore the Coaching That Clicks Custom, a BoodleBox bot.
The Notice-Wonder-Next Protocol
When analyzing video with teachers, use this simple three-step protocol:
1. NOTICE: What objective data do you see? Some items to look for include:
Student talk time,
Wait time,
Questioning patterns
2. WONDER: What questions does this raise? Those can include:
Why did engagement drop during independent work?
How might different grouping affect participation?
3. NEXT: What will you try differently? The goal is to provide feedback that is specific and actionable for the next lesson)
This protocol keeps feedback descriptive rather than evaluative, empowering teachers to drive their own growth.
A Quick Reminder
Remember, video-based coaching is not about catching mistakes, assessment, or staff appraisals. Rather, it’s about making student learning visible and focusing on finding ways to apply evidence-based strategies. Adopt structured protocols like EIIR and PRISM, or whatever protocols you already have in place, to work towards better student outcomes. With a 0.99 effect size, video-based coaching represents one of the most powerful levers we have for improving instruction.
Helpful Links
PDFs (video-based coaching strategy)
Coaching Cycles (PDF) — coaching-cycle workflow using reflections + video to document growth and structure coach/teacher conversations.
Need to convert an MP4 video file to an MP3 audio file with drag-n-drop? Or, resize and compress an image that’s too big to upload? Need to convert PDFs into a single merged, text file in MarkDown format for easy upload to a Gen AI chatbot? Or, perhaps, you have a series of photos you want to convert into a slideshow? If you use technology with any frequency, you may find yourself struggling to find the right, no-cost apps to get the job done. I can’t tell you how many times I’ve installed a “free” program only to find that there are in-app purchases needed. But there is an easier way.
Generative AI Makes the Command Line Easy
Much of what you pay app creators for is available for free, if only you knew how how to use the command line on Windows (or GNU/Linux and Mac). The problem is, finding the right combination of command line options for a program like FFMPEG or ImageMagick can take a few Google searches and hours of experimentation. I have put in that time, and then, about six months ago, a solution so obvious I mentally kicked myself for not trying it sooner.
Note: While all the examples in this blog entry focus on .bat files, you can easily ask your Gen AI chatbot for a BASH file (filename extension is .sh). Bash files work a little differently on GNU/Linux, and that would necessitate a separate blog entry. This idea works for all sorts of challenges, including moving fonts from Windows to GNU/Linux, as one person found out when I put together this tutorial for them using Gen AI, too. The point is, give it a go.
You already know what the solution is, right? I decided to ask Generative AI to give me the right command line options for shrinking a series of videos. You see, I had recorded a host of my daughter’s wedding videos, and they had gigantic file sizes. While I work with video all the time, babysitting video conversion programs to do the conversion (which takes forever) was not on my dance card. When Perplexity AI gave me the right combination for a bat file, I realized I might be on to something. A half year later, I’m now using AI-authored BAT (Windows) and BASH (Mac and Linux) to do all sorts of tasks.
What’s a BAT/BASH file?
I learned about BAT files in my freshman college class, “An Introduction to MS-DOS.” My professor, back in the late 1980s, had me write BAT files by hand. At that time, the realization that coding was not my thing sank in a little deeper.
BAT/BASH files are like recipes. They are a list of instructions that instruct your computer to do things automatically. Instead of you clicking through menus and typing commands one by one, the script does it all for you. Think of these as a to-do list for the operating system. The .bat file is for the Windows operating system. On Mac and GNU/Linux, the recipe of instructions file is known as a .bash file. They do the same kind of thing, only they use language that each operating system understands.
Some real world examples you can use script files for include:
Automatic back up of files from one folder to another every day (great for backing up that USB drive you carry around with you)
Launching multiple programs at once with a single click
Renaming hundreds of files in seconds instead of one by one (ugh, this is a pain)
Clean up of temporary files
Now, my “Aha!” moment came when working with large video and image files I wanted to share with others. Others do not need the high quality video or image. A smaller, more shrunken version I can fit inside of an email or text message attachment will do just fine.
Sample Prompt for a BAT File: Make Animated GIF from JPGs
“What is the simplest way to create an animated GIF from an existing JPEG/JPG/PNG? The JPG has a character I need to animate. I’m looking for a bat file” (Source)
To that end, I set up a ChatGPT Project (you can do this in BoodleBox, Claude, Copilot, and Gemini, too) that I labeled “Technical.” Then I added some custom instructions to it to guide my work. These custom instructions take the hard work out, and instead, make it easy to drag-n-drop files on top of a .bat file on Windows to get the desired action.
One Example: PDF/Txt/MD to MD File Merge
Problem: Need to merge PDFs, Txt, or Markdown (MD) files into a single file?
There are limits on how many files you can upload to a knowledge bank (e.g. ChatGPT limits you to 25 files for a Project/CustomGPT). Or, perhaps you want to optimize your text files before putting them into a knowledge bank.
It will convert from PDF/Text/Markdown (MD) formatted files to markdown, then merge the MD files into one. This means instead of giving your GenAI twenty-five different PDF documents, you provide only one. This takes up less space AND the Gen AI chatbot processes the text file faster. You can always rename the “merged.md” to “merged.txt” if your AI chatbot of choice can’t handle markdown.Google Gem and BoodleBox Bots can produce markdown files, but won’t accept them as input unless you change the filename extension to txt.
Free Software to Install for Windows
You will need to install some software, all of it free, open source (FOSS). Then, you will need to add this software to the PATH so it can work anywhere you happen to be (such as your Desktop, in some deep directory/folder on your device). Let’s look at both steps now, which you will only need to do once.
Install Software
You will not need all the software I have in my tools folder (more on that in a moment), but you can start with the following software:
Install ffmpeg and ensure ffmpeg.exe is on PATH.
Install ImageMagick and ensure magick.exe is on PATH.
Optional: install ExifTool (exiftool.exe on PATH) for metadata stripping.
Optional: install Whisper CLI (whisper on PATH) for subtitles.
A convenient step by step is available to assist you in installing the software programs above. That will enable you to take advantage of solutions such as the ones detailed below:
Not sure how to add programs to the PATH on Windows 11? Let’s walk through that now.
Add Programs to the PATH
The way I do this is to put all the EXE files for the programs I installed (see list above) and put them into a single folder called “tools.” Then, I put this tools folder at the root level on my computer. You can find the root folder by going to your “My Computer,” double-clicking Windows (C:) and you will see a list of folders. This is where you can create a “tools” folder and/or drag your Tools folder there.
Here’s an excerpt of a video tutorial I made for a colleague. It shows you how to add a folder called “tools” (like the one shown above) to the PATH on Windows 11.
The main benefit of this? I don’t have to keep modifying my “Environment Variables,” that is, adding items to the PATH again and again. Instead, I simply put the new program into the C:\tools folder and it’s ready to go.
How BAT Files Work with Drag-n-Drop
Here are a few of the problems that I’ve asked ChatGPT to create bat files for me for:
Automatic media conversion to MP3/OGG/MP4. This bat file prompts me for the desired output format I want then gives me a smaller file in that format.
Batch image resizing and compression. This bat file relies on ImageMagick, compresses the image file, and then saves it with a new name.
Convert PDFs + text/Markdown into a single merged Markdown document. This bat file takes advantage of Xpdf tools to create a merged text file.
Here’s a demonstration video showing one in action:
Although I have set up various .bat files, for this blog entry, I asked ChatGPT to generate a menu of options file. This addresses a variety of needs you may run into (or at the least, that anyone working with video, images, and PDFs runs into from time to time).
Get the BAT file and custom instructions for mgConvertGPT, a ChatGPT Custom GPT I made that you can simply use. Of course, you can take all the info I have shared and make your own Project or Custom GPT, BoodleBox Bot, or Gemini Gem.
You can ask it to help you create a BAT file, as you can see in the screenshot below:
Simply click the “Copy code” button, open Notepad on your Windows computer, then paste in the code. Save it with a filename like “Compress_Video_For_Sharing.bat” into your C:\tools folder. Then, drag-n-drop video files on top of the bat file. You can drag-n-drop a single video file, several, or a folder’s worth.
Here’s what that will look like when you drop a video file on it:
Advance organizers prepare the brain for new learning. They give students a structure before content appears. This simple shift can transform how learners encode and retrieve information. Instead of treating new material as isolated facts, students slot ideas into existing frameworks. The result is deeper learning with less cognitive strain.
More on Easing Cognitive Load
When students use tools that help them get ready to learn, their brains do not have to work as hard to make sense of new ideas. They already have a place to put the information. This means they can understand more, remember more, and stay focused because their minds are not getting overloaded.
An Advance Organizer for This Blog Entry
Before we go any further, allow me to share an advance organizer. It reflects the content of this blog entry. Feel free to print it out (PDF) or draw your own on a piece of paper using the image below as a suggestion.
Why Advance Organizers Work
Research supports what teachers often notice first. John Hattie places advance organizers at an effect size of 0.41, above the point at which learning benefits become significant (d=0.40). That effect size speaks to what happens inside the brain. You understand new information by linking it to something you already know. When that connection is weak or missing, learners struggle. Advance organizers build those bridges before instruction begins.
A few other benefits to consider appear below. Advanced organizers:
Activate prior knowledge.
Reduce cognitive load
Give students a sense of purpose before content appears
Provided a structure in mind, learners enter a lesson ready to process, not just receive
Seven Advance Organizers You Can Use Tomorrow
The following organizers work across grade levels and content areas. Each one meets learners where they are and moves them toward deeper understanding. Which organizers are you NOT using and might consider to use in the future?
1. Prequestions: Priming Curiosity
Prequestions invite students to think before learning begins. These questions do not require full answers. Their power lies in prompting learners to search their memories and anticipate what matters.
Example: Before a unit on ecosystems, ask: “What do living things need to survive in their environment?” Students begin making connections before the first slide appears.
2. Expository Organizers: The Overview
These offer a brief, structured summary of what students will learn. They provide the big picture, allowing learners to see how pieces fit together.
Example: Prior to teaching algebraic equations, outline how equations represent balanced quantities. This frames the details that follow.
3. Narrative Organizers: Story as Scaffold
Stories give unfamiliar ideas familiar shape. They place abstract concepts within relatable experiences. The emotional connection strengthens recall.
Example: Start a Great Depression lesson with a short personal story from the era. Students step into the historical moment rather than study it from afar.
4. Skimming and Previewing: Strategic Scanning
Previewing teaches learners to recognize structure before diving into text. They examine headings, diagrams, graphs, and bolded terms. This primes them for comprehension and helps them anticipate key concepts.
Example: Ask students to skim a chapter for three minutes and note what they believe the central ideas will be. This simple act heightens attention during reading.
5. Analogies and Metaphors: Familiar Bridges to New Concepts
One way to connect the new to the familiar relies on analogies and metaphors. These help students grasp abstract ideas by connecting them to existing knowledge.
Example: When teaching about electrical circuits, compare them to water flowing through pipes. Voltage might be described as water pressure and current as the flow rate. This makes abstract concepts (e.g. electrical terms) more concrete.
6. K-W-L Charts: Purposeful Knowledge Activation
K-W-L (also KWHL) charts remain one of the most effective ways to capture what students bring to a lesson. The K and W columns serve as the advance organizer. They reveal what learners already know and what they hope to uncover.
Example: Before studying ancient civilizations, have students complete the K and W sections. This directs attention and drives inquiry.
7. Quadrant Outlining: Structured Thinking
Quadrant outlining breaks information into four essential parts. This reduces overload and supports analysis. It works especially well before writing tasks or complex discussions.
Example: Have students outline an argumentative essay using four boxes: main idea, supporting details, connections, and summary statements. The structure clarifies their thinking before drafting.
Technology Tools That Support Advance Organizers
Digital tools expand what is possible and increase student ownership.
The key is choosing tools that make thinking visible without slowing learning. You might even consider handwritten digital tools, like Rocketbook’s smart reusable notebook paper (if that’s an affordable option).
Why Advance Organizers Work
Advance organizers mirror how our brains constructs meaning. They activate prior knowledge and lessen cognitive load. They assist students in getting ready to learn. When implemented in a consistent manner, they have a positive impact in how students connect, reason, and remember.
Note: This blog has been updated with fresh content on 12/31/2025.
“I had an argument with my son today about learning math and coding. He told me he doesn’t need to learn them anymore because he can just use Google’s AI mode to get homework answers and build anything he wants,” said Ebrahim K. His post made me ask the following:
If coding and essay writing are no longer the vehicles for teaching critical thinking, logic, and design, then what physical activity could be relied upon instead?
In this blog entry, I take a stab at figuring that out. And, it turns out that I’m not the only one to ask the question.
The Core Question
“How do we develop critical thinking skills in ways that GenAI can’t reproduce?” The question pushes me to leave my office, to step outside. GenAI has impacted everything from ethical systems to global economies. Now, it’s coming for critical thinking and student cognition.
Embodied learning may be one possibility. It is a tough one given our preference for traditional learning.
Definition: Embodied Learning Embodied learning uses movement and physical activity to help students learn and remember better. It’s an educational model focused on using the body during lessons to stimulate involvement and improve the learning process (source).
Why Embodied Learning Matters Now
“Our most basic form of learning relies on the body’s knowledge. Our most basic form of learning in childhood is preverbal. However traditional schooling forces us to check our bodies at the door,” says R.L. Lawrence.
What if physical presence and how we experience the world with our bodies help reconnect us with learning in ways that GenAI nullifies? Our bodies facilitate genuine connections with others. These connections and experiences fuel our learning. While GenAI may be our digital friend, nothing replaces a handshake or hug.
Aside: It’s difficult to imagine school staff now having to deal with students involved with AI companions. That’s one more distraction from left field…
Could students learn math through building actual structures? Run classroom businesses with real products? Or, perhaps, create physical displays that respond to human movement? The goal does not focus on GenAI usage, but seeks to encourage building solutions in the real world. These are multi–sensoryapproaches that develop students’ neural pathways, not train GenAI models.
The Power of Physical Thinking
For many teachers, critical thinking emerges through writing or coding. The goal is to have students externalize their thinking, making it visible. But as GenAI tools make fabricating student work child’s play, teachers wonder: “What can students do instead to make their thinking visible?” Perhaps, what projects can engage students?
Embodied learning requires real-time adaptation and attention. Failing to pay attention has physical consequences. What’s more, students must employ spoken rhetoric to persuade and/or inform others.
Building Through Physical Experience
Consider pottery-making: students receive instant tactile feedback about cause and effect. They must be aware of pressures, temperatures, and techniques—a sensory dialogue that teaches problem-solving in ways GenAI cannot replicate.
Other hands-on activities that develop critical thinking include:
Blacksmithing: Reading color changes to determine temperature
Gardening: Understanding subtle, esoteric knowledge about plants and soil
Woodworking: Learning material properties through direct manipulation
Rock-climbing: Problem-solving routes with real physical consequences
Each activity involves developing critical thinking through physical interaction with materials. These activities involve students in discovering rules and knowledge. And, it must all be learned through doing.
The PRISM Framework in Action
The PRISM framework (Patterns, Reasoning, Ideas, Situation, Methods) offers a way to design embodied learning experiences. In a wilderness trek, students:
Patterns: Identify natural indicators of direction and weather
Reasoning: Analyze how patterns create survival strategies
Ideas: Generate multiple solutions to challenges
Situation: Evaluate which solutions fit circumstances
Methods: Implement and test approaches with real-world feedback
This progression moves students from surface to deep learning, ending with transfer. This is something rarely achieved in traditional settings, according to one colleague I spoke with a few weeks ago.
AI as Partner, Not Replacement
Rather than fighting AI, educators can design embodied learning experiences where GenAI enhances the process. In school gardens, students might:
Generate initial design ideas with GenAI, then physically implement and adapt plans
Brainstorm solutions collaboratively, then apply ideas in real interactions
Analyze data from physical problem-solving experiences
In this way, real learning happens as students work to physically implement solutions co-developed by humans with GenAI.
The Enhanced Makerspace
In embodied learning situations, students cannot offer AI’s answers as final solutions. The physical experience of being in proximity to others, working with real materials, prevents that. As GenAI infiltrates classrooms, we must move back toward physical, problem-solving spaces. Consider an adaptation of Peggy Reimers’ work on makerspaces:
PRISM Component
Traditional Making
Embodied Learning Enhancement
Gen AI Integration
Patterns
Students identify patterns in materials, tools, and construction techniques through hands-on exploration
Physical manipulation reveals material properties (wood grain, metal flexibility, cardboard strength) through tactile feedback
AI analyzes project failures/successes to suggest material combinations and structural patterns
Reasoning
Trial and error with physical materials teaches cause-effect relationships
Body knowledge develops through repeated physical actions (measuring, cutting, assembling) with immediate sensory feedback
Students query AI for initial designs, then must physically test and adapt based on real-world constraints
Ideas
Brainstorming and sketching project concepts
Physical prototyping with found materials allows rapid iteration and collaborative building
AI generates multiple design variations; students physically build and compare outcomes
Situation
Adapting projects based on available materials and tools
Real-time problem-solving when materials break, measurements are wrong, or tools malfunction
AI provides troubleshooting suggestions that students must physically implement and verify
Methods
Following maker processes: Think→Question→Design→Create→ Struggle→Enjoy→Collaborate→Try→Solve→Fail→ Problem Solve→Reflect→Learn
AI documents the making process, but students must physically demonstrate mastery through successful builds
New Possibilities for Embodied Learning
The more GenAI finds its way into our work and classrooms, I have to wonder how we might move back from unnatural, artificial settings into more physical, problem-solving and learning spaces. What do you think? Will all of us begin the slow transition to more physical, embodied learning?
While working in a small, suburban school district as a technology director, I found myself dragged into several Professional Learning Committee (PLC) meetings, but only a few seemed to prove real results. I couldn’t help but wonder, “What questions should we really be asking about our data? How could these critical questions change how a PLC approaches student learning?” That’s why I’ve developed these five critical questions for Texas educators to consider.
Focusing on What Matters: The TEKS Mastery Conversation
These five questions can easily become a framework that lends some structure to PLC meetings. Use these questions to focus conversations on ensuring students achieve essential learning standards. Let’s take a look at each of the questions below.
Question 1: Which TEKS have students gained proficiency in?
This first question cuts straight to the heart of our work as educators: “Which TEKS and/or state/local standards have we ensured students gain proficiency in during this instructional cycle?”
What’s powerful about this question is that it focuses on what student have actually learned—on specific TEKS that students have gained proficiency in. The benefit is that it shifts the conversation from curriculum coverage to student outcomes. As you might imagine, this question aligns perfectly with the the SOLO Taxonomy’s relational level, where students connect ideas within the content area.
Think of this in the context of an example for fourth grade math. A fourth grade team meets and identifies that 85% of students had mastered TEKS 4.4A (add and subtract whole numbers and decimals). They discover that only 62% had mastered 4.4H (solve with fluency one- and two-step problems involving multiplication and division). If you are familiar with the PRISM Framework, you may think of this as a way of recognizing patterns in the data.
Question 2: What evidence do you have?
This question focuses on teachers showing evidence of student learning. It requires proof of the assertion that students showed proficiency, “What evidence (interim assessments, district benchmarks, CBAs, or formative data) demonstrates student mastery of these essential TEKS?”
It’s essential to proportion claims to the evidence available. This evidence could include:
District benchmark results
Common formative assessments
Student work samples
Exit ticket data
Digital tool analytics (Quizizz, Pear Assessment, etc.)
Using the PRISM framework, this question helps us move from simply recognizing patterns in data to reasoning about what those patterns mean for instruction.
Question 3: Which students need interventions?
The third question personalizes our data: “Which specific students interventions within our MTSS framework to master grade-level TEKS and readiness standards?”
This question transforms abstract percentages into actual students with names and faces. The goal is to identify students by name, and ensuring you are able to identify growth opportunities for them. One approach you can take advantage of is the Amazing Lesson Design Outline (ALDO). It recommends conducting pre-assessments to discover what phase of learning students are in. MTSS covers both academic and behavioral supports in a holistic, coordinated system for all students
Question 4: What interventions will we put in place?
This question is intended to shift us towards action. It requires us to consider interventions and supports we can implement as a team. It asks: “What targeted interventions and supports will our team implement within the MTSS framework to ensure these students achieve mastery of essential TEKS?”
As you may know, MTSS merges RTI and PBIS. It provides a unified framework that addresses academic, behavioral, and social-emotional supports. At the same time, it encourages collaboration between educators and stakeholders.
Using evidence-based strategies, we determine exactly how we’ll provide additional time and support. This might include:
Small group instruction during designated RTI time
Digital tools like Padlet’s AI features for collaborative learning
Reciprocal teaching strategies for reading comprehension
Flexible grouping based on specific skill deficits
Question 5: How can we differentiate or enrich learning activities?
The final question ensures we’re meeting all students’ needs: “What GT differentiation or enrichment activities will we provide for students who have demonstrated mastery of grade-level TEKS and readiness standards?”
Too often, we focus exclusively on struggling students. This can mean that high achieving students are left bored, isolated on their own plateau of learning. Our purpose in answering this question is to ensure we are planning for extension activities. That is, activities that push students to the extended abstract level of the SOLO Taxonomy.
Making It Work in Your School
To get this done, you need to leverage your entire team. Five steps to get you started appear below:
Schedule regular data meetings. Set aside dedicated time (bi-weekly works well) for teams to analyze evidence and plan interventions.
Create evidence templates. Develop simple forms for teams to document their evidence and intervention plans.
Use digital tools wisely. Leverage technology like Poll Everywhere or Mentimeter for quick formative assessments that provide real-time data.
Build intervention time into your master schedule. Without dedicated time, interventions rarely happen consistently.
Celebrate growth. Recognize both student progress and teacher effectiveness in moving students toward mastery.
The goal is to create a culture of collective responsibility and ensure a focus on students gaining proficiency in essential standards.
Have you ever wished for a teaching assistant who never sleeps, always has fresh ideas, and can help your students develop critical thinking skills? If you have, you may find that using these examples of custom GPTs (also known as Gems or Bots depending on the tool you use) as assistants that can lighten your load. And, to model how you can use ANY chatbot as a personal assistant in your classroom, I’ve organized some instructions for you that you can get a copy of and try in your own preferred Gen AI chatbot.
Let’s explore seven powerful AI assistants that can change how you and your students approach critical thinking, writing, and content creation. They are examples of ways you can create your own custom GPTs and bots. Not sure you know what GPTs are? See the Glossary at the end of this blog entry.
#1 – Building Digital Literacy with the SIFT Approach Bot
If you’ve read many of my blogs here at TCEA TechNotes, you know that misinformation and information literacy are critical. Students need structured frameworks to evaluate online content in a critical manner. Mike Caufield’s SIFT Method offers one viable approach. The SIFT Approach Bot guides learners through a four-step process that transforms how they consume digital information.
If you are in the habit of accepting online sources at face value, The SIFT Approach Bot will teach you to ask thoughtful questions about authorship and context. “I never realized how my emotional reaction to a headline was clouding my judgment!” I heard one participant say. It’s true, though. Doom-scrolling and reacting often begins with an emotional reaction.
The bot walks learners through the steps of the SIFT Method:
Stop and check emotional reactions to content
Investigate the source by examining authors and publishers
Find better coverage through corroborating sources
Trace claims back to their original context
What makes this tool particularly effective is its visual interface with clear “Questions to Ask” at each stage, making critical thinking accessible to learners at various levels.
#2 – Deepening Analytical Thinking with the PRISM Framework
As an educator, you know firsthand how hard it is to get students to move beyond surface-level understanding. The PRISM Framework Bot(used under your guidance with students) addresses this challenge by structuring thinking across five areas:
Patterns: “What patterns do you notice in this information?”
Reasoning: “How do these elements connect logically?”
Ideas: “What creative connections can you make?”
Situation: “How does context influence this topic?”
Methods: “How can we validate these conclusions?”
This framework aligns with educational progression models like SOLO Taxonomy (in fact, it’s based on SOLO). It helps you and your students shift from identifying basic facts to making complex connections and transferring knowledge to new situations.
Try PRISM in a middle school science class, or social studies class. You will notice students progressing from listing observable characteristics to making sophisticated hypotheses about cause and effect. That is a valuable skill to learn and the PRISM Framework Bot can assist with that.
#3 – Transforming Meetings with the Outline Helper Bot
This is a custom GPT I use several times a week. How many meetings have you left wondering, “What did we actually decide?” The Outline Helper Bot transforms chaotic meeting transcripts into clear, hierarchical summaries that drive action and accountability. I often audio record meetings, then run the transcript through the Outline Helper Bot.
This simple tool creates structured outlines using:
Hierarchical Structure: Roman numerals for main topics, letters for subtopics, numbers for details
Action Item Organization: Groups next steps by responsible party with clear deliverables
Clean Formatting: Uses Markdown for professional, scannable output
What I truly appreciate is how it takes the tedious work of formatting my notes into an organized outline. It makes tracking progress and sharing notes after a meeting so much easier.
#4 – Organizing Complex Information with Quadrant Note-Taking
Remember teaching Cornell Notes? The Quadrant Note-Taking Bot takes structured note-taking to the next level by guiding students to organize information into four strategic sections:
Main ideas
Supporting details
Personal connections
Summary/synthesis
This approach makes complex topics more accessible while supporting writing development across grade levels. It’s a great way to synthesize information and organize it quickly. I like to use it after I’ve taken notes on content that I need to remember but need to see in a fresh, new way.
#5 – Supporting Academic Integrity with the AI Citation Assistant
At the recent TCEA 2025 AI for Educators Conference, someone referred someone to my blog entry, Citing AI Assistance: Five Approaches to Try. Realizing I had not revisited the topic since, I decided to take another look. You know, it is quite important to model for students how to use these tools in an ethical manner. The AI Citation Assistant ensures teachers support students in maintaining academic integrity while leveraging AI’s power.
This practical tool:
Generates citations in multiple formats (MLA, APA, Chicago)
Explains when and how to cite AI tools
Provides examples of ethical AI use in academic contexts
Recommends citation tools like MyBib and Monica
Proper attribution applies not just to human authors but to AI-generated content as well. This tool is offered in that spirit of assistance.
#6 – Building Language Through the Vocabulary Enhancement Bot
Tired of vocabulary flashcards? The Vocabulary Enhancement Bot tries to assist you in creating engaging, multimodal activities:
Word Poetry Slams: Challenges students to create poems using target vocabulary
Stop-Motion Animation: Guides students to visualize word meanings through frame-by-frame animation
Interactive Word Games: Reinforces understanding through gamified activities
What I appreciate most about this bot is how it differentiates activities by grade level, offering simpler rhymes for K-2 students, haikus for grades 3-5, and more complex poetry for grades 6-12.
#7 – Unlocking Writer’s Block with the Freewriting Coach Bot
The blank page can be a show-stopper for some beginning writers. You have, no doubt, experienced students freeze or groan in frustration at the blank page. The Freewriting Coach Bot can help you assist students in overcoming writing anxiety through structured exercises that get ideas flowing.
This practical coach:
Guides students through timed writing sessions
Provides prompts based on images or topics
Teaches the FLOW method: Fast & Timed, Loose, Open, Write
Helps silence the internal editor that blocks creativity
The goal is to help students realize the blank page is an opportunity, not a problem. It’s also a way to quiet the murmurings of the internal editor, while building writing fluency. You may find my blog entry on freewriting of interest, as well.
Creating Your Own Custom GPTs for Education
What makes these tools particularly powerful is that they’re just the beginning. As an educator, you can design your own custom GPTs tailored to your specific classroom needs. Consider creating GPTs that:
Guide students through your specific writing process
Provide scaffolded support for complex projects
Offer differentiated feedback based on student needs
Support specific content areas with specialized knowledge
The process of creating a custom GPT is straightforward. Start with a clear description of what you want the GPT to do, provide examples of ideal responses, and refine through testing. Many educators find that their custom GPTs become invaluable classroom assistants that extend their teaching presence. To get you started, I’ve provided the custom instructions for all the GPTs in this blog entry. Feel free to steal and adapt them for your own use.
Glossary
Note: Definitions and chart generated by ChatGPT
⚙️ Core Terms
GPT — A custom ChatGPT with tailored instructions, files, and behavior.
GPT Store — Directory of public custom GPTs (like an app store).
Main Instructions — Core persona and behavior prompt.
Knowledge — Uploaded files or documents that a GPT can refer to.
Actions — API-based tools your GPT can use to interact with other systems.
Do Not List — Rules to prevent unwanted GPT behavior.
🤖 BoodleBox-Specific Terms
BoodleBox Bot — A custom chatbot in BoodleBox using multiple AI models.
@mention — Way to tag and summon bots inside a chat.
Quick Commands — One-click prompts that launch bots or workflows.
Knowledge Bank — BoodleBox’s file system for bot knowledge.
Instructions Wizard Pro — A tool to design detailed instructions (paid feature).
💎 Gem (Generalized Builder Term)
Gem — A no-code AI bot (e.g., in Google Gemini), like a simplified GPT.
Gem Wizard — Step-by-step bot creator with minimal input needed.
One of my favorite poems is Dromgoole’s The Bridge Builder. Part of my admiration for the bridge builder stems, not only from his selfless act, but also that he has the skill to build a bridge. Today, you and I both have access to powerful tools that enable us to build our own bridges.
Any educator starting with GenAI may be experiencing a bit of tool paralysis. But there are skills you can learn that make you effective—you simply have to choose which tool works best for you and use it effectively as an assistant in your problem solving process.
Avoiding Tool Paralysis
Choice boards are big, aren’t they? Students love them because they give them choice, an option to self-differentiate their own learning activities. But, put a human being in front of a display of tools and options, and our brains shut down. Which do we choose? Many of us are facing GenAI tool paralysis now, as these tools appear practically everywhere online. Even our computer operating systems and smartphones are sporting these technologies. If you are feeling paralyzed by the overwhelming amount of choices, the best path forward is to adopt a practical, privacy-first approach.
Start Here: Privacy as Your North Star
Before we dive into the approach, let’s address the elephant in the room: student data privacy. Here’s a quick checklist you can use:
Your Privacy Checklist: ✓ Who owns the data from AI interactions? ✓ How is student information protected? ✓ What permissions are required? ✓ Is there an educational privacy policy?
Use these basics with other tools, like TCEA’s PROTECT rubric (also available as a custom GPT) as the baseline for determining a tool’s safety.
Now, let’s take a look at a critical concept: the PRISM Framework. If I had to teach you one thing, it would be this. You can use it again and again.
The PRISM Framework: Your Decision-Making Compass
Before implementing any AI solution, run it through PRISM framework:
Patterns: What works in classrooms like yours?
Reasoning: Does this align with your pedagogical beliefs?
Ideas: How might this enhance (not replace) your teaching?
Situation: Which specific challenge does this address?
Methods: How will you measure success?
A simple framework that takes you from simple to complex use of GenAI tools in your classroom environment, school, or organization.
Context Is Everything
You may already know that GenAI tools are only as smart as their training. What you may not realize is that these tools that allow you to provide MORE context (such as BoodleBox Bots, ChatGPT Custom GPTs, Claude Projects, Google Gemini Gems). Think of it like giving directions with examples. The more specific you are, the better the outcome.
Note: In fact, I shared about this topic at my session, Building Knowledge Stacks, at the TCEA AI for Educators Conference (it may be too late to view the video, but you can see my session resources with cross Gen AI tool video tutorials). Want a Zoom or F2F professional development session? Reach out to me via mguhlin@tcea.org.
Your Context-Building Toolkit
To recap some of the key ideas, when building your custom Bot, GPT, or Gem (using BoodleBox, ChatGPT, or Google, respectively), consider gathering these key pieces. Please note that the instructions’ real example are shortened to keep this chart from getting overwhelmingly long.
What to Provide
Why It Matters
Real Example
Custom Instructions
Shapes every AI response
“Create 5th-grade reading questions using SOLO Taxonomy—start with multistructural, build to extended abstract”
Knowledge Bank
Grounds AI in your reality
Upload your actual rubrics, reading lists, and curriculum maps
Learning Goals
Keeps AI on target
“Help students move from identifying facts to synthesizing concepts”
Pro Tip: Spend 10 minutes uploading your key documents before your first AI conversation. This investment will pay dividends in relevant, aligned outputs. More importantly, it will save you tons of back and forth prompting trying to get the GenAI tool to understand what you want.
Transform AI from Tool to Teaching Partner
Often, people see GenAI as a vending machine—a quick way to short-circuit the tedium of work. That’s because so many see the time-saving benefits of GenAI. But with GenAI, like with carpentry, you want to measure twice before cutting. You’ll also need to make sure your materials are readily available, your tools are sharp, and that there’s a pencil tucked behind your ear.
Instead of treating AI like a vending machine, treat it like a collaboration partner or force multiplier. One way to do that is to use a process or method like the one below:
The 4-Step Collaboration Method
1. Start with Clarity Instead of: “Make me a reading activity” Try: “I need differentiated reading activities for my 3rd graders at three readiness levels.” I have a curated prompting workbook that I share with folks I provide professional learning to. But there are ample resources online for prompting help.
2. Show, Don’t Just Tell Share an example: “Here’s an activity that worked well last week. Notice how it scaffolds from basic recall to critical thinking….” By explaining the result you want up front, you improve the probability you will get what you want the GenAI to put together.
3. Co-Create Your Formula Ask: “Based on our conversation, can you write instructions that would help you create similar activities in the future?” This is the quick way to get the GenAI to make your next series of similar tasks easier. Now that you’ve gone through the process, you can have it build or customize your instructions to reflect that refined process.
4. Save Your Success Turn those co-created instructions into a reusable template (future you will be grateful that you did). The way to turn those instructions into reusable template is simply to save them somewhere. Some people put them in Google Keep, a Google Doc, OneNote, or Joplin notebook.
Your Next Move
The path forward isn’t about using AI for everything or trying to learn every tool. Instead, analyze your problems and ask, “Will GenAI make a difference here?” Then, start with that one challenge, whether it’s differentiating reading materials or creating formative assessments. Remember to always put privacy first, build a rich context for your GenAI tool of choice. Collaborate and ask the chatbot to assist you. End with documenting your journey, and sharing it with others. Whatever you choose, remember:
Pick one tool, one task, and one hour. The future is for the bot builders—those who can look at the obstacles they face and engineer the tools they need to get the work done. I imagine building a bot to solve a particular problem rather than provide the entire solution—a critical component of a bigger process. What could you do if you understood how to build a bridge to span the divide?
—on specific TEKS that students have gained proficiency in. The benefit is that it shifts the conversation from curriculum coverage to student outcomes. As you might imagine, this question aligns perfectly with the the SOLO Taxonomy’s relational level, where students connect ideas within the content area.
