New agent mode can iterate on its own code, infer tasks not specifed, recognize errors, and fix them automatically, GitHub said.

GitHub Copilot may be the single most useful AI companion I've used, but it's about to get a lot better with agents and other improvements.

Here, too, GitHub Copilot succeeded where Microsoft Copilot failed. The challenge here is that I'm testing the AI's ability to create a script that knows about coding in AppleScript, the Chrome object model, and a little Mac-only third-party coding utility called Keyboard Maestro.

New agent mode can iterate on its own code, infer tasks not specifed, recognize errors, and fix them automatically, GitHub said.