Can the presence of a URL in an LLM prompt influence the output based on the content of that URL?
I think, yes! But it took me weeks of experimentation and I learnt a lot about what User Agents do and what data is in the models
I was a little unsure when Jev could be used, and this (and Yusuke's route demo earlier) is probably one of the first and clearest usecases that made it just click.
A Hono middleware with Jev on Workers AI. One yes/no question in, a probability out, block if it's high.
Same 15 lines, different question:
- Block spam
- Stop prompt injection before it reaches the LLM
- Hand angry customers to a human
I am here to predict the (partial) death of your favorite JavaScript frameworks and libraries…
In the age of agentic coding, I seriously doubt that frameworks that provide a lot of implicit magic that helped humans be productive will be the successful frameworks. IMO, code
If you are interested, I built github.com/paulkinlan/fau… many months ago.. plugin in any model. It generates images and videos too so you can even fully browse a hallucinated YouTube!
you can just hallucinate the entire internet with Qwen 3.8 27b running at 2,000 tokens/second?
part 2 of turning @cerebras + @Alibaba_Qwen 3.8 27B into an OS:
built an offline browser with zero network calls and mounted it directly the JIT ubuntu desktop.
no wifi. no scraping.
I swear we need an `<input type=secret>` along side `<input type=password>`, heck maybe just `name=secret`
I'm getting tired of API keys being saved like passwords.