Clement Tsang
e5f13d17fc
Adding this as I've been getting some AI interaction lately - while I'm fine with it, I think now is a good time to add some guidelines at a repo level.
2.0 KiB
AI Policy
The use of AI (i.e., LLMs) for coding is welcome. However, there is a high bar for any contribution to this project, and as such, contributors are responsible for any code they publish.
AI should not be used to entirely generate comments when communicating with maintainers. We expect comments on our projects to be written by humans. Comments we believe are entirely AI-generated may be hidden without notice.
If you are opening an issue, we expect you to at least describe the problem in your own words. Issues must still follow the specified template. Issues that are unproductive and appear to be entirely AI-driven may be closed without notice.
If you are opening a pull request, you are expected to explain the proposed changes in your own words, as well as to respond to questions. You are also expected to have tested the changes if applicable, and follow the PR template and checklist. Do not copy responses from AI when replying to questions from maintainers. Pull requests that violate these polices may be closed without notice.
If you wish to include context from an interaction with AI in your comments, it must be in a quote block (e.g., using >)
and disclosed as such. It must be accompanied by human commentary explaining the relevance and implications of the context.
Do not share long snippets/walls of text.
It is also acknowledged that AI is very useful for communicating if one is not fluent in English. If you are using AI to edit your comments for this purpose (e.g. grammar, spelling), please take the time to make it so it still reflects your own voice and ideas. If you're using AI tooling for translation purposes, then it's recommended to write in your native language and include the translation in a quote block.
This policy was inspired by ripgrep's AI policy and uv's AI policy.