miniblog.

One subtle behaviour of Claude that wasn't obvious to me: whilst each conversation is transient, permissions persist across conversations. So if you've given permission to run e.g. 'cargo test' or even 'cargo run', you need to be sure that all future invocations are safe too. You can see the current permissions with /permissions.

It's so hard to talk about probabilities. Casual conversation often conflates 'the likelihood of an event happening' with 'how strongly I believe in my assessment'. For example, I'm very confident (say 90%) that the coin in my pocket is 50% likely to land on heads.

Having a conversation with a PDF is a really fun application of LLM technology: https://www.chatpdf.com/