Re:Skills Officially Comes to Codex
by JamesSwift on Saturday December 20, 2025 @04:17PM
(#46339715)
How would you evaluate it if the agent were not a fuzzy logic machine?
The issue isnt the LLM, its that verification is actually the hard part. In any case, its typically called “evals” and you can probably craft a test harness to evaluate these if you think about it hard enough
The issue isnt the LLM, its that verification is actually the hard part. In any case, its typically called “evals” and you can probably craft a test harness to evaluate these if you think about it hard enough