Can agents be proud of their work?

In 2022, we got better output from large language models by asking them to show their work. In 2026, can we get better output from autonomous agents by asking them to be proud of their work?