Typical Claude response:
"Yes, you're right, but you don't even realize the complexity of the problem and the hard part is not where you'd expect"
Snarky. Following the script from "Brilliant but Cruel" to appear smarter and more trustworthy sciencedirect.com/science/articl…
Now it's much easier to scale up your agent evals in skillgrade!
‣ Split eval.yaml across files
‣ Task metadata & filtering
‣ Agent model selection
‣ Real-time progress tracking in the CLI
Thanks to Greg Radzio for the great work!
github.com/mgechev/skillg…
