xpnsec

People simply get that pelican dude to ride a bike. If that sucker’s feet touch the pedals, we know that the model is good for six more weeks.

Out of interest, what does your local setup look like? Is this a spare GPU you’re using or have you got something more designed for local LLMs?

Where they are very useful is as a documentation source for LLMs. For example, I work in infosec and often have to reference DSLs (Cobalt Strike aggressor script for example). Having a skill which is an offline index to carved up function docs, which an LLM can use without having to think, then search for, then download huge 1 page documents with all function documentation, and pollute the context… very useful.

“In addition to disabling the camera on devices when tampering is detected, we work across Meta to remove ads, posts, and Marketplace listings that advertise these kinds of tampering services and we will take action, up to banning accounts that do this.”

Disable our glasses LED.. straight to jail! Tamper with elections and influence votes across a country… meh.

Love this idea!

As someone with a couple of code bits tattooed on my body, if anyone is planning on more than a few lines of text, speak to your tattoo artist first.

I once wanted a long bit of code on my arm. I thought it was a good idea, but speaking to my tattoo artist, he explained that after a few years, the text would likely blur and merge together until it was difficult to read.

Ended up going with a smaller snippet which is still clear to this day (15 years later). And seen some others who didn’t listen to advice with an illegible blob where their clean code one was xD

It’s like I asked the exact question :) Your experience is certainly not unique. I see the stories, and often feel like it is just a skill issue on my part. So I ask for advice to help correct my workflow, and I receive a mixture of suggestions on getting the LLM to produce maintainable code (which is always welcomed when it helps to improve the output). But then I noticed that there is a small pocket of people who tell you that you are wrong, or it is a skill issue, or their view is completely at odds to yours. That they have been 10x’ing their output for months. So you ask if they can share a codebase or GitHub link which demonstrates how they have managed to tame the LLM on larger projects, and people go silent.

So now I’m trying to let the code do the talking as one method of learning. Hunting through GitHub looking for SDD projects and trying to understand what works vs what is parroted on X.

Love that idea of compounding your skills and driving that into a new area!

I think Windsurf is to Antigravity as Atom was to VSCode

More interestingly, what frameworks/harnesses/architecture are people using to drive multi-agent workflows?