Open Codenames: a new way to benchmark LLMs

Open Codenames is a fresh benchmark framework for LLMs that enables quick, task‑specific model comparisons. To adopt it, manually review the README, dependencie