Note also the "Claude Capybara" reference in the undercover prompt: https://github.com/chatgptprojects/claude-code/blob/642c7f94...
This seems like a good way to weed out models: ask them to include the term capybara in their commit messages
At least this was known with the Mythos "early blog post" fiasco.
This seems like a good way to weed out models: ask them to include the term capybara in their commit messages