AI Terms for Vibe Coding

Add an AI assistant to my website that people can chat with.

This AI travel guide sounds confident, but I do not want to trust the opening hours. Research the official website or reservation page online and include the source link and update date.

I can't code. I just want to tell an AI what I need and have it build it for me.

Can AI understand my screenshot too, not just the text I send it?

As the task gets longer, the AI starts missing a newly confirmed requirement. What should the next turn include?

Everyone says AI pricing is based on tokens. What is a token, and how does it relate to what I type?

The AI forgets what I said at the start of a long chat. Does it have a memory limit?

Where do I set rules for the AI, like its tone and what it can or can't answer?

The earlier messages are still visible in the chat. Why did the AI edit the wrong second item?

Build a client quote list: client name, amount, date, and status, stored in the browser, newest first.

The table is still visible in the chat. Why does the AI not know what to do when I send only “continue”?

AI gives me a big block of text I can't use directly. Can it return the answer in a fixed format?

Show the answer as it comes in, like someone typing, instead of making me wait for the whole thing.

I don't want to direct every step. Can I give it a goal and let it figure out how to finish it?

The same AI works well sometimes but repeats the same mistake elsewhere. What can I change besides the model?

The AI can research the itinerary and flights, but it must ask me before paying or issuing a ticket.

Can I split a big task among a few helpers at the same time, then get one combined result?

AI can talk, but how do I get it to actually check the database or make a request?

Can the AI check the result after each step before deciding what to do next?

After a check fails, should the AI continue editing, stop to ask me, or end the task?

What do Provider and Model mean in this configuration, and where will my request actually go?

I've switched the model API endpoint to our team's self-hosted proxy. Where do I update the Base URL to make it work?

Requests started working after I switched to a relay URL. Check who can see the prompts, files, and key along that path.

My AI tool only has settings for OpenAI, but I want to use DeepSeek. Can I just swap the Base URL and API Key?

Connect the LLM API for me; read the API Key from environment variables, and never hardcode its value in the source code.

Help me update the default model in the AI tool's config file to DeepSeek, and increase the timeout limit to 60 seconds.

Configure two profiles for me: 'work' using our company's private API, and 'personal' using my own API Key. Show me how to switch between them.

I keep reminding the AI to use TypeScript and our existing styles. Can these requirements live in project rules?

Does this tool support a post-write Hook? If so, have it format only the files that just changed.

AI plugins, hooks, and Skills all look like extensions. Which one actually fits this need?

Can the AI connect to our issue tracker and tell me who owns #1024?

I have to teach the AI the same process every time. Can it remember how to do this task?

Routine changes inside this project can continue automatically, but ask before deletion, wider access, or external actions. Which permission setting fits?

Both sandboxing and command confirmation are meant to protect my computer. What does each one limit?

Why do some AI tools take forever to show the first word while others answer right away?

My API bill blew up this month. What am I actually paying for?

I'm getting an error because I called the API too often. What does that limit mean?