A key is writing a clear spec ahead of time, which provides context to the agent as it works in the codebase. (View Highlight)
Having a document for the agent that outlines the project’s structure and how to run e.g. builds and linters is helpful. (View Highlight)
Asking the agent to perform a code review on its own work is surprisingly fruitful. (View Highlight)
Finally, I have a personal “global” agent guide describing best practices for agents to follow, specifying things like problem-solving approach, use of TDD, etc. (This file is listed near the end of this post.) (View Highlight)
