Organize PRs by category
I was speaking on 2025-05-20 for the policy team, but I thought I should be open about it because there seems to be some misconceptions in the world about ranking people by the number of likes they have.
2025-05-23
nishio We have a working code that labels the PRs that come into Team Mirai's policy once an hour and puts them into a Markdown file. If you paste it into ChatGPT, you can chat with AI based on what kind of PRs are coming in that category.
https://github.com/team-mirai/policy-pr-data/tree/main/label_reports
2025-05-22
nishio I created a system that crawls PRs for Team Mirai's policy every hour and keeps the JSON containing all the data updated with the latest information. By instructing Devin in Japanese, I was able to extract the PRs with specific conditions from all PRs in about 2 minutes.
Devin's use in improving Team Mirai policy
2025-05-20
Press this button to copy and
https://gyazo.com/2f1d24014eb70df4eb1fcb535a73ae9b
If you paste it into ChatGPT and write, for example, "a quick summary of what it says", you get results like this
https://gyazo.com/f3a229bac244d176e5f6a6254d29f1aa
https://chatgpt.com/share/682b6017-82ac-8011-b409-2935c56e83ae
I recommend o3 instead of 4o.
As for the problem of "there are too many things to do and I don't know where to start," I think that consulting with o3 like this will solve the problem.
I'm sorry to inform you, but this method is not available since 200 cases have come in the README.
Team Mirai 2025-05
---
This page is auto-translated from /nishio/PRをカテゴリごとにまとめる using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.