The resulting script is available here, and it worked first try to scrape up to 20,000 videos (the max limit). The resulting Python script has very Pythonic code quality following the copious rules provided by the AGENTS.md, and it’s more robust than my old script from 2021. It is most definitely not the type of output I encountered with Sonnet 4.5. There was a minor issue however: the logging is implemented naively such that the API key is leaked in the console. I added a rule to AGENTS.md but really this is the YouTube API’s fault for encouraging API keys as parameters in a GET request.
Card trading — Trade Pokémon cards with other players and get expert tips on how to play from event leaders.
，详情可参考im钱包官方下载
Зеленский собрался изменить дипломатическую командуЗеленский: Будут усилены некоторые направления дипломатической работы в Европе，更多细节参见heLLoword翻译官方下载
I then added a few more personal preferences and suggested tools from my previous failures working with agents in Python: use uv and .venv instead of the base Python installation, use polars instead of pandas for data manipulation, only store secrets/API keys/passwords in .env while ensuring .env is in .gitignore, etc. Most of these constraints don’t tell the agent what to do, but how to do it. In general, adding a rule to my AGENTS.md whenever I encounter a fundamental behavior I don’t like has been very effective. For example, agents love using unnecessary emoji which I hate, so I added a rule: