Code Generation Prompts #122
Description
On the accompanying blog post you write:
A surprising amount of the system's behavior comes down to how we prompt the agents.
Getting them to coordinate well, avoid pathological behaviors, and maintain focus over long periods required extensive experimentation.
The harness and models matter, but the prompts matter more.
Can you please release the prompts you used ?
Or even better: Write a followup article about what prompts worked better than others, and potentially why ?