Welcome to Day 5! Now that we know what agents are, let's look at how we can use them in our day-to-day lvies. Today we'll address a universal pain point: planning travel. This feels particularly relevant for me today as it's frigid in New York and I'm dreaming of a warm weather getaway.

We've had tools like Google Flights alerts for years. They help but you still end up doing a lot of the work: setting alerts for multiple routes, filtering for airlines you prefer, ensuring nonstop options, and constantly monitoring all of these different options. Agents can do all of that for you.

🧠 What You’ll Learn Today

• How agents can help tackle real-world logistics
• A quick experiment to try for your next vacation

🎯 How and Why to Use an Agent to Save Time

Agents tool access and reasoning capabilities make them more adaptable than most out-of-the-box alerting tools.

Sticking with a travel example, agents can monitor:

• Several airports (LGA/JFK/EWR → SAV + HHH)
• Specific airlines and ticket classes (Delta forever ♥)
• Direct flights only (kids + layovers = no thank you)
• Travel windows instead of specific dates, things like days of the week in a specific range (leave Friday, return Monday in Feb or March)
• Target price range

And they send you an update when there is a match. You're now spending just minutes a week instead of hours on tracking travel options.

🛠️ Hands-On Experiment: Track a Spring Break Flight

A favorite tool of mine for this type of agent is Gobii. Head there and create a “Digital Worker” with a prompt like:

Help me track flights for a spring break trip from NYC to Orlando.
Here are my constraints:
• Delta flights only
• Nonstop
• Depart between April 12–15
• Return April 19–22
Track nearby airports too (TPA + MCO).
Alert me if prices drop below $300.

Gobii will take this information and build a detailed prompt that executes on the task you’ve assigned to the agent. Tomorrow, we'll dig deeper into what's going on under the hood here.

✍️ Recap

• Agents automate ongoing, multi-step, real-world tasks
• They rely on preferences, context & judgment to make decisions, not just preset filters

📬 Keep the Streak Going

Next up: we’ll take a deeper look at the agent created during today’s experiment and see what we can learn from it.

👉 Subscribe at 12DaysofAgents.com to get the next edition

👉 Forward Day 5 to a friend who loves a good deal (or hates layovers) 😄