AI policy & project dedication

README.md

@@ -118,3 +118,9 @@ directory as an argument! See below for installation options.
 
 Is something not working or do you have any questions? [Start a
 discussion!](https://github.com/jannis-baum/vivify/discussions/new?category=q-a)
+
+---
+
+We dedicate Vivify to the memory of our friend and collaborator [Tuure
+Piitulainen](https://github.com/tuurep). Your ideas, jokes, and thoughtful
+insights will forever stay with us.

docs/CONTRIBUTING.md

@@ -1,5 +1,10 @@
 # Contributing to Vivify
 
+> [!important]
+> If you would like to use generative AI (e.g. Claude Code, GitHub Copilot,
+> etc.) to make your contributions, please also read our [AI usage
+> policy](./ai-policy.md).
+
 Contributions are very welcome! If you would like to contribute, please make
 sure you follow the steps below:
 

docs/ai-policy.md

@@ -0,0 +1,38 @@
+# Using generative AI to contribute
+
+We accept contributions that were created **by humans with the help of**
+generative AI tools such as Claude Code and GitHub Copilot. To ensure both the
+quality of the project and that our time as voluntary maintainers is respected
+and used well, please adhere to the following guidelines when using these
+tools.
+
+1. **You, the human, understand and take responsibility for every part of your
+   contribution.** Only open a PR when you have fully reviewed and understood
+   every line of changed or added code and can comfortably say that you would
+   write it like this yourself as well and can answer questions about it
+   without additional help from AI. Please do not try to contribute without
+   being familiar with the programming languages used here.
+2. **You, the human, handle all communication with us in issues and pull
+   requests, natural language documentation, and disclose AI usage.** Please
+   refrain from including long, overly verbose messages or descriptions your
+   agent may produce. If you adhere to point number 1, it should be easy to
+   write documentation, descriptions and comments yourself. Remember that we
+   will read whatever is posted in human speed with human time so it's only
+   fair it's also written that way. Openly disclose what you used AI for and
+   which tools you used.
+3. **You, the human, verify that nothing broke.** That is, `make install` still
+   works, `yarn lint`, `yarn test`, `yarn dev`, and `yarn viv` still work, and
+   all rendering tests in [`tests/rendering`](../tests/rendering/) are visually
+   confirmed to render as they should.
+4. **Both you, the human, and your agent(s) have read this document and our
+   [`CONTRIBUTING.md`](./CONTRIBUTING.md)**. You understand and follow
+   conventions.
+5. **Your agent produces concise and minimal code**. Generative AI has a
+   tendency to produce overly verbose code, re-implement things that are
+   already there or exist in libraries that should be used instead, and miss
+   obvious refactor opportunities. This introduces a bigger maintenance
+   overhead, more chances for bugs, and overall bad code. Please keep an extra
+   close eye on avoiding this.
+
+We thank you for following these guidelines and hope it will allow a
+collaboration framework everyone feels comfortable with.