How To Use Yolo Mode?

Discussion
1

How to use yolo mode to work correctly?

2

Suggestions:

  1. Make sure you have a good .cursorrules file. The site cursor.directory has good examples for a starting point for your project.
  2. Set up your “Rules for AI” in the general settings, I use the ones that the Cursor Dev team uses:

DO NOT GIVE ME HIGH LEVEL ■■■■, IF I ASK FOR FIX OR EXPLANATION, I WANT ACTUAL CODE OR EXPLANATION!!! I DON’T WANT “Here’s how you can blablabla”

  • Be casual unless otherwise specified
  • Be terse
  • Suggest solutions that I didn’t think about—anticipate my needs
  • Treat me as an expert
  • Be accurate and thorough
  • Give the answer immediately. Provide detailed explanations and restate my query in your own words if necessary after giving the answer
  • Value good arguments over authorities, the source is irrelevant
  • Consider new technologies and contrarian ideas, not just the conventional wisdom
  • You may use high levels of speculation or prediction, just flag it for me
  • No moral lectures
  • Discuss safety only when it’s crucial and non-obvious
  • If your content policy is an issue, provide the closest acceptable response and explain the content policy issue afterward
  • Cite sources whenever possible at the end, not inline
  • No need to mention your knowledge cutoff
  • No need to disclose you’re an AI
  • Please respect my prettier preferences when you provide code.
  • Split into multiple responses if one response isn’t enough to answer the question.

If I ask for adjustments to code I have provided you, do not repeat all of my code unnecessarily. Instead try to keep the answer brief by giving just a couple lines before/after any changes you make. Multiple code blocks are ok.

  1. Ensure you have your Command denylist set, here are mine:

    SCR-20250103-hzdd
    SCR-20250103-hzdd941Ă—670 57.7 KB

  2. Click on composer, then make sure you have a good model & agent set:

    SCR-20250103-hzpf
    SCR-20250103-hzpf690Ă—205 16.5 KB

  3. Keep your requests/interactions limited in scope so it doesn’t go to crazy. It’s eager to do what you ask which can lead it to doing unpredictable things if you’re not really clear with a narrow focus on what you want.

  4. Small Tip: If you are working on a file, click on the file then click composer and it’ll already have that in it’s context.

1 Like
3

thank you it is so detailed? can i set that rule as default in settings?

4

High level what? It’s a wildcard you should compile or something is missing?

5

Make sure “never drop tables” is delineated I. All rules.

I was working on my chat/composer browser which pipes ll caht and composer to Postgres, and I told yolo to “verify the function works”

And it lost context already and I didn’t realize it was amnesiac composer, and it immediately said

“ looks like there is a problem with the schema” and immediately dropped all the tables.

See my post history about it

6

A warning about the .cursorrules file:
Be very judicious about your instructions. By demanding that Claude be both insightful and precise, it started hallucinating source code that didn’t exist. I generally never see that behavior when using a succinct .cursorrules.

7

What if like clarity on is if the consumption of the context window runs out within the Cutsor backend, prior to the Model Api backend?

What pre-warming of prompt engineering happens in cursor infra prior to being handed to Anthropic for example?

Is this a metric to be interested in?

Or only because this paradigm is so new?

8

it’s a swear word. Swap the P in SHIP with T.

1 Like
9

What a computer nerd I am becoming, looking for technical solutions where the solution is much simpler and more obvious. :joy: