Conversation
|
|
||
| Never treat this as a "normal" Laravel e-commerce package: catalog data is read straight from Magento's database schema via Eloquent, not through Magento's own ORM or REST API — GraphQL is only used for write/transactional actions like cart and checkout. | ||
|
|
||
| IMPORTANT!! For all Rapidez specific questions please check the https://docs.rapidez.io/llms.txt and the relevant links first! |
There was a problem hiding this comment.
Is this really the best way to make sure any LLM understands this? Is there a standardized way to give "important" context? Isn't everything in this file important enough anyway?
Given the sliding context windows issue, if it's really important it should also probably be all the way at the bottom of the file as a precaution. Especially if we expand this list below eventually.
There was a problem hiding this comment.
- Not sure, I've seen other packages do it and it seem to work fine.
- For the second part; I don't understand what you're saying here.
| - Catalog, category, CMS data, etc everything that's we need to read from the Magento → Eloquent models querying the Magento database directly (see `Rapidez\Core\Models\*`) | ||
| - Cart, checkout, customer mutations, etc everything that has authentication, is dynamic or need to write to Magento → Magento GraphQL. |
There was a problem hiding this comment.
These 2 instructions don't seem particularly clear to me and I could easily see an LLM tripping up and mixing them up. I'm not an expert on this, but I would personally expect a small table to be easier to interpret.
There was a problem hiding this comment.
So far so good, suggestions are welcome
The
CLAUDE.mdhere is just in case you're working directly on the core. The guideline is for Laravel Boost, currently there are no skills. I did create some but it was kind of copy of the docs which didn't felt very maintainable. We could sync the docs automatically to skills but lets see how this is going with a link to https://docs.rapidez.io/llms.txt which was added in: rapidez/docs#127. Also added a new AI page in the docs: rapidez/docs#128