Robert Cedeno used to spend more than an hour writing 30 home tour voiceovers. With a couple of AI tools, he gets them done in 6 to 7 minutes.
Agents hear a lot about AI saving time. What’s harder to find is a clear look at how working agents set it up and what they use it for.
Over the past six months, BAM contributors and podcast guests have shared how they’re using AI to research local stories, vet content ideas, write scripts, manage follow-up, automate repetitive tasks and even audit whether their business plans actually fit into their calendars.
The tools vary, and so do the use cases. But the common thread proves that the agents getting the most out of AI aren’t just asking it questions. They’re building workflows around work they already do.
Here are 10 of those workflows you can steal.
Script and Video Prep Workflows That Cut Production Time
These two agents use AI to prep their videos before they hit record.
1. Write home tour voiceovers in minutes
Robert Cedeno runs a 25-agent team in Southern California and posts 10 times a day across three Instagram accounts. He explained how he keeps up that pace when he joined The Broke Agent on a recent podcast.
He cloned his voice on ElevenLabs, an AI tool that can clone a user’s voice and turn text into audio using that voice. From there, he set up a prompt in Meta’s Muse, an agentic AI assistant, to write his voiceover scripts.
Cedeno walked through the instructions he gave Muse.
“All I do is I set up the prompt, like, ‘Hey, this is how I want the voiceovers. Anytime I give you an address, I need you to create three different voiceovers. One, I need to start with something like, this is what 850,000 gets you in this area…'”
Muse pulls the property details from Zillow and writes the three versions. ElevenLabs records them in his voice, and his editing team adds the audio to the home tour videos.
Muse schedules his calls and responds to his DMs too, and he approves those replies before they go out.
2. Build a Claude Skill to vet content ideas
Real estate content creator Dave Hutch joined The Broke Agent to talk about a Claude Skill he built to test his content ideas before he films them. A Claude Skill is a saved set of instructions Claude follows each time you use it, and he made his with Claude’s Skill Creator tool.
The skill runs each idea past a five-seat board of advisors:
- Two marketer seats modeled on Alex Hormozi and Sabri Suby
- A Champion who pushes the idea toward its best version
- A Skeptic who tries to kill it
- A Chair who makes the final call
The board argues it out, and the Chair decides whether the idea moves forward. Hutch writes the scripts himself once an idea passes.
He made a point of clarifying how easy this is to set up.
“Remember, I want to preface this, but you don’t need to be techy to do this. The whole point of AI is to make you more efficient.”
AI Research That Runs in the Background
These two agents have Claude watch their local sources so they don’t have to check them by hand.
3. Schedules MLS searches with Claude
Hutch has a second Claude workflow, which he demonstrated in an Instagram reel. He uses the Claude extension for Chrome, which lets Claude work inside your web browser, to run an MLS search on a schedule.
View this post on Instagram
Here’s how he sets it up:
- Open the MLS and launch Claude from the browser
- Give it a specific instruction, like finding condos under $500,000 in his farm area with a price drop in the last three days
- Click the three dots on the result and select “convert to task”
- Schedule it to run each morning
The same setup works for other searches:
- Weekly pulls of new listings or price changes in a farm area
- Price drop searches built around a buyer’s price range and area
- Daily checks for CRM leads nobody has touched in more than 48 hours
The hours that went into pulling those lists by hand can go toward appointments and closings.
Hutch said the extension can take on more than MLS searches.
“Once you have the Claude extension installed, you can literally ask it to do any of your repetitive tasks and get it done for you while you work on things that make you more money.”
4. Set up Claude to comb through city filings
Katie Day leads the #MOVEMETOTX team at REAL and built her Houston Unlocked Instagram account to 46,500 followers by covering Houston news and new developments. She broke down her research system in a conversation with The Broke Agent.
She uses ChatGPT and Claude for research and scripting. Claude scans Houston city council websites for her and sends back a summary of what’s been filed.
Day explained what she has Claude looking for.
“I have it searching the Houston city council websites for planning agendas and plat filings and different stuff like that and giving me kind of a summary of the things that are happening because there’s so many different things that are being filed and talked about and most of them aren’t newsworthy, but I want it to kind of comb through them and also remind me when the meetings are so that I can go look at the physical agendas themselves just cuz, again, AI might miss something that’s actually a big deal to the people of Houston.”
She keeps the story picks for herself and reads the original sources. Once she knows what she wants to cover, she gives a tool bullet points from a news article or city council update and asks for a script draft. From there, she edits it into her own voice.
Agents outside Houston can point the same kind of setup at their own city council or county permit office. The BAM post lays out the steps:
- Find the pages where your city posts its council agendas, planning and zoning commission agendas, plat filings and subdivision filing reports
- Check whether your city uses Legistar, Granicus, CivicPlus or a similar platform, since these often have search or RSS feeds built in
- Write one clear prompt for the scheduled task that says what to check and what counts as new since the last run
- Spell out what you care about in the prompt, like specific districts, zoning types, project names or dollar thresholds
- Set the schedule to match how often those sources post
- Pick where the summary goes, like your email or a doc you’ll open
- Run it for a couple of cycles and compare the results to the real agendas before you trust it
Teaching AI Who You Are
These workflows help AI learn your voice and your business, and the last one helps AI recommend you to buyers and sellers.
5. Train Claude Projects on his own content
Chris Smith, best-selling author of The Conversion Code and co-founder of Ella, Beacon and KnwnLocal, gave the opening presentation at BAM Fest 2026. He told agents to master Claude and walked through how he uses Claude Projects.
A Claude Project is a workspace where you upload files for Claude to reference. Smith uploads his past content, and Claude learns his voice, tone, style and formatting from it.
He described what happens once that’s in place.
“Once you have that, you can literally say, give me five tweets. And every tweet it gives me is on brand. They know that I like to be contrarian. They know that I like to be funny. They know that I like to use stats.”
He uses Claude Artifacts to build working tools like content generators and calculators. He uses Claude Design to make on-brand graphics from his existing visual assets.
6. Voice brain dump
BAM co-founder The Broke Agent wrote about how he stopped re-explaining his business every time he opens a new chat with Claude.
He talks for 10 to 15 minutes, using voice mode or a recording app, and covers:
- His professional background
- His current role
- His team and tools
- Projects he’s working on
- His career stage
- His target clients
- The markets he works in
- His business goals
He pastes the transcript into Claude and asks it to add the key details to its memory. To check the result, he asks Claude to summarize what it knows about him, and he refreshes it each month.
He described what changed after he set it up.
“Now when I open Claude, it already has the context for everything I’m working on. No re-explaining. No starting from scratch. It just knows.”
7. Get recommended by AI
Real estate marketing coach Will Draper sat down with The Broke Agent to explain how tools like ChatGPT, Claude, Perplexity and Gemini decide which agents to recommend. He breaks it into a five-part framework:
- Foundation
- Authority
- Relevance
- Proof
- Media
Foundation comes first, and it starts with using the same name and brokerage on every profile online. He recommends writing one positioning statement that names who you help, in three lengths:
- One sentence, for your elevator pitch
- 750 characters, to match the Google Business Profile limit and reuse across directories
- 1,500 characters, for the about page on your website
From there, he suggests building case studies around two or three real specialties. He’s started writing review drafts for his clients to copy and paste, so the reviews include specific details.
His clients have seen results:
- A Windermere, FL agent generated an estimated $20 million in production from ChatGPT-driven leads in about two months
- A case study about a client relocating from the UK led to a $5 million listing appointment within two weeks
- A Houston agent got a listing appointment after ChatGPT quoted his Google Business Profile back to a prospect word for word
Draper said agents need to change the question they’re asking.
“I think people should stop asking ‘How [do] I rank in ChatGPT?’ Stop asking how do I rank in it? You want to ask what would ChatGPT need to know about me to actually recommend me?”
Emails, Follow-Up and Daily Admin
These two workflows take repeat emails and follow-up off an agent’s plate.
8. Run your daily tasks
Haley Ingram, founder and CEO of Coffee & Contracts, uses Claude every day in her business. She walked The Broke Agent through Cowork, a feature in the paid version of Claude that can log in to your tools and do tasks for you. The paid version costs $20 a month.
Here’s what Cowork can do for an agent, according to Ingram:
- Build a morning dashboard that pulls from your email, calendar, CRM and project management tool
- Run that dashboard at a set time each morning with no input from you
- Time-block your calendar
- Fill Canva templates with local details and your photos for hyperlocal content
- Draft emails into your inbox
- Send a recap at the end of the day showing where your hours went
She sets up repeat tasks as Skills, the same feature Dave Hutch used for his board of advisors. Here’s how she builds one:
- Do the task with Claude, going back and forth until the output is right
- Tell Claude to save it as a Skill
- Next time, type a backslash and pick the Skill, or describe the task in plain words
- Schedule the Skill to run on its own at a set time, if you want
According to the BAM post, the scheduling step is where the time savings come from.
Ingram explained how a Skill works for a weekly email newsletter.
“A Skill is essentially like a standard operating procedure or guidelines that you are giving Claude that anytime that you mention this specific task, it’ll run that Skill automatically.
“So, for example, if you have an email newsletter that you send out every week, instead of every time you go into Claude saying, ‘Hey, I want to write my email newsletter, here’s the details, here’s what we’re doing.’ You just say, ‘I need to write the email newsletter.'”
9. Hands tasks to a team of AI agents
At BAM BBQ 2026, Phil Stringer showed how he runs his business with a team of AI agents. An AI agent is an AI tool that can carry out tasks on its own, and each of his is trained for one job.
His starter team has four agents:
- Director of Marketing
- Database and Referral Manager
- Content and Listing Manager
- Reputation and AI Visibility Manager
Stringer explained how a task gets to the right agent.
“You have a task, you send it to a manager or a supervisor agent or a director of ops, whatever you want to call it. And based on that task, it says, ‘Well, we have a sub-agent who specializes in that thing.’”
He gave an example from his own business. One prompt, “let’s run a new live training,” kicked off the follow-up for the whole event:
- Personalized promo emails scheduled through his CRM
- A reminder sequence that goes out the day before, the morning of, an hour before and when doors open
- Text and Instagram DM reminders for people who opted in
- Follow-up for no-shows with the replay and a deadline
An AI Audit of the Weekly Calendar
10. Have AI check your plan against your calendar
Katie Lucie used AI to find out whether her business plan fit inside her week. She built a task list for each of her lead pillars and put the full list and her existing time blocks into AI for an audit.
It flagged:
- Commitments she hadn’t set aside time for
- Recurring tasks she’d left out
- More output than her calendar could support
Lucie described what the exercise did for her.
“That exercise was incredibly clarifying. It’s very easy to write ‘social media’ on a business plan. It’s much harder, and much more useful, to prove your calendar can produce the six pieces of content you just promised yourself you were going to create every week.”
Keep a Real Person Behind the Content
Several of these agents draw a line on what they hand off to AI. Katie Day picks her own stories, and Dave Hutch writes his own scripts.
In May, The Broke Agent wrote a warning for agents who type in a prompt and post whatever comes out. He made clear he’s not against AI and uses it all the time himself.
He described what 100% AI-generated content looks like to the people scrolling past it.
“You can REALLY tell in video scripts because they’re completely lifeless. It’s turning people into content robots who take a breaking hyperlocal news story, throw it into ChatGPT, and spit it back out with zero opinion or personality of their own.”
Your audience wants to see a real person behind the post, with stories and opinions of your own.
He closed with a test to run before you post.
“Moral of the story: think before you post. Is this actually adding anything? Is this something only YOU would say? If not, that’s where the danger is.”
Ultimately, your audience (and potential clients) want to see there’s a real person with a life story and opinions behind your content.
Nothing wrong with using AI to reclaim time for income-producing activities. Just make sure you’re not using it in a way that alienates the very people you’re trying to reach.
