I’m having awesome success working with recursive agents. I discussed my experience with them. [0] > Claude's attention doesn't distinguish between "instructio

I’m having awesome success working with recursive agents. I discussed my experience with them. [0] > Claude's attention doesn't distinguish between "instructions I'm writing" and "instructions I'm following" -- they're both just tokens in context. It takes a little human help in the first iterations but after a while it will start to iterate and improve unsupervised. [0] https://github.com/adam-s/agent-tuning https://github.com/adam-s/agent-tuning