In this solo episode, Heather explores one of the biggest shifts happening in the AI space right now: the move away from relying on prompt libraries and toward building systems powered by context.

A couple of years ago, prompt engineering was everything. People were collecting prompt packs, saving templates, and searching for the “perfect prompt” that would magically produce the right result. And at the time that made sense. Early AI tools needed much more structure, narrower inputs, and a lot more hand-holding.

But things have changed.

Today’s AI tools are better at understanding natural language, asking follow-up questions, inferring structure, and working inside persistent environments like projects, custom assistants, and knowledge bases. That means the real advantage no longer comes from having a huge library of prompts, it comes from giving AI the right context.

Heather unpacks why prompt libraries are becoming less central, what should replace them, and how short-term rental professionals can start building smarter AI systems that actually reflect their business, voice, standards, and guest experience.

If you’ve been wondering why AI sometimes still feels generic, even when you use “good prompts,” this episode will help you see what’s missing - and what to build next.
‍

In this episode, Heather talks about:

• Why prompt libraries made sense in the early days of ChatGPT and other AI tools
• The limitations of treating prompts like one-size-fits-all solutions
• Why modern AI tools now need less perfect wording and more meaningful context
• The difference between a one-off prompt and a true AI-powered system
• How projects, custom assistants, and knowledge bases change the way we work with AI
• What a context library is and why it matters more than a prompt library
• The five layers of context engineering:
  
  • task context
  • business context
  • audience context
  • quality context
  • resource context
• How short-term rental professionals can train AI to better reflect their voice, policies, and guest expectations
• Why skills and workflows help create more consistent AI-supported results across a team
• What practical assets to start building now if you want AI to become a real operational advantage
  ‍

Key takeaways

Heather’s central message is simple: prompts still matter, but context matters more.

A saved prompt might help you get started, but it won’t know your business. It won’t understand your tone, your brand standards, your guest profile, or how you handle tricky situations. That’s why even a strong prompt can still produce output that feels generic or slightly off.

What makes AI genuinely useful is the system behind the prompt.

That system might include a project workspace, a custom assistant with a clear role, a knowledge base full of your business materials, and a context library that teaches the AI how your business works. Add repeatable workflows on top of that, and you move from experimentation to real operational support.

Heather also shares a practical mindset shift: stop asking only, “What should I type?” and start asking, “What does the AI need to know to do this well?”

That’s the heart of context engineering.
‍

Memorable lines from this episode

“The prompt is no longer the product. The system around the prompt is the product.”

“Your prompts are the steering wheel. Your context is the engine.”

“Prompt engineering taught us how to ask. Context engineering teaches us how to prepare.”

“A prompt library is a toolbox. A context library is an operating system.”
‍

Practical next steps

If you want to apply what Heather shares in this episode, start small:

• Create a simple business overview AI can reference
• Write down your brand voice and communication style
• Build a guest or audience profile
• Gather your best documents into a reusable knowledge base
• Keep a small set of starter prompts for the tasks you do most often
• Turn your common tasks into step-by-step workflows AI can follow

You do not need to build everything at once. The goal is to start creating reusable context that improves every future interaction.
‍

Mentioned in this episode

• Lovable
• Claude
• ChatGPT
• Wispr Flow
• AI Ambassador Programme
  ‍

Final thought

This episode is a timely reminder that AI success doesn’t come from collecting more prompts. It comes from building better foundations.

For short-term rental professionals, that means creating systems that teach AI about your business, your guests, your standards, and your voice. Once you do that, AI stops being a novelty and starts becoming a real extension of your operation.