Skip to content

dvelton/ask-bigger

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Ask Bigger

See what's possible with AI tools that can take action.

Two categories of AI tool

Most people's experience with AI is conversational. ChatGPT, Claude, Gemini in their default chat modes: you type, they respond. These tools are genuinely useful for writing, analysis, brainstorming, and explanation. Millions of people use them well every day.

But they operate entirely within the chat window. They can't touch your files, run code, search your email, or take any real-world action. When you need something done, they can advise you on how to do it. The doing is still yours.

A newer generation of tools works differently. GitHub Copilot in the terminal, ChatGPT with tools enabled, Claude with computer use: these read and write files on your machine. They run shell commands. They search your email and calendar. They build working software and deploy it. They create documents, analyze datasets, interact with APIs and databases.

This isn't a better chatbot. It's a different category of tool, and it changes what's worth asking for.

The expectations gap

If you've mostly used conversational AI, your sense of what's worth asking for is calibrated to what those tools can do. That calibration is accurate for chatbots. But it can also carry over to agentic tools, where the ceiling is much higher.

Someone who's only used conversational AI might ask an agent to "summarize these meeting notes." The agent can do that, but it can also extract every action item, draft follow-up emails to each attendee, and propose the next meeting's agenda. The first request is fine. The second gets more value from the same interaction.

This is the pattern Ask Bigger is built around. Not that people are doing something wrong, but that there's often more available than people realize, and seeing a few examples tends to shift your sense of what's worth trying.

The compounding effect

Once you see an AI agent do something you hadn't considered asking for, it tends to open up adjacent questions. Could it also handle X? What about Y? People who start exploring agentic tools often describe a compounding effect where each new capability they discover changes what they think to try next.

The tool matters. But the more interesting shift is in your own sense of what's possible.

What this site does

Ask Bigger shows what becomes available when you move from conversational AI to AI agents.

Each example has three parts:

  • A common starting point (how most people would approach this with a chatbot)
  • What you could ask instead (the fuller version, sized to what an agent can actually do)
  • A concrete prompt you can copy and paste into an AI agent right now

Browse by category or search by task. The examples cover writing, research, data analysis, operations, technical work, legal, and strategy.

Running locally

No build step. Open index.html in a browser.

Contributing

Add examples to data.js. Each entry needs a category, a title, the small ask, the bigger ask, a concrete prompt, and a set of search keywords. Keep the prompts copy-paste ready with [bracketed placeholders] for the parts users need to customize.

About

See the gap between what you typically ask AI tools to do and what's actually possible.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors