OpenInterpretability adopts the Contributor Covenant v2.1.
- Be kind. Interpretability is a young field; many contributors are students or first-time researchers. Kindness is a feature, not a constraint.
- Be honest. Report what actually happened in your experiment. Honest negatives are welcome; spin is not.
- Be helpful. If you see someone struggling in Discussions, on an issue, or in a PR review — say the thing you wish someone had said to you.
- Assume good faith. If a PR or comment could be read two ways, pick the generous reading.
Harassment, demeaning language, doxing, or any form of discrimination based on identity. Intentional bad-faith technical content (fabricated benchmark numbers, plagiarized notebooks, off-topic self-promotion).
Report issues to hi@openinterp.org or DM a maintainer on GitHub. Reports are confidential. Maintainers will respond within 72h with one of: no action needed, private warning, temporary ban, permanent ban. Patterns matter more than one-off incidents.
This Code applies to all spaces tied to OpenInterpretability: issues, PRs, Discussions, our Discord (when it exists), mentions on X tagging @openinterp, and in-person events where the project is represented.
Adopted 2026-04-23. Updates to this document require a PR and two-maintainer approval.
Full text: https://www.contributor-covenant.org/version/2/1/code_of_conduct/