Recent content by Taylor J

It would help to remember that LLM's are just statistics under the hood and are all setup to not disappoint the end user I.E: If an llm can't find you a correct answer to your question they'll give you any answer as to never come back with their hands empty, unless specified that you are okay...

That onus is on us, but we are lucky enough that in 2.3.10 the team provided us with a new php command to init all needed agent/claude.md files.

They seem to be running that beta per what @briansol said.

Was not expecting the phrase tool so much that I actually had started on one for myself with claude just yesterday lmfao. Nice work on the agentic coding additions as well, will be adding those to what I already have going which should give me a nice boost in non hallucinations which tbh I'm...

Taylor J updated [TaylorJ] Blogs with a new update entry: 1.6.1 Read the rest of this update entry...

That's on the user profile view right? Will have a fix out this weekend for that.

You don't even bother to have it write in a first person pov so that it tries to sound like it's you saying this and not Gemini.

Taylor J updated [TaylorJ] Blogs with a new update entry: 1.6.0 Read the rest of this update entry...

Update should hopefully be out tonight. Had some preoccupation with life situations and spent my time bug squashing and getting unit tests finally setup after fixing/adding the above two solutions.

Hm, I may try bringing over the Railsy way of being able to seed programmatically or maybe create a PHP version of Factory Bots where entities would be able to be mocked/faked entirely without side-effects.

Do you have any recommendations / your own way of doing integration tests with XenForo? Or would you be waiting until we have SQLite support?

It would be nice to see a docs nav item in the forums for easier access.

There is one that @Chris D has that works pretty well for the short time I sold a premium addon.

I can get behind this.