diff --git a/skills/review/SKILL.md b/skills/review/SKILL.md
new file mode 100644
index 0000000..95ea9c8
--- /dev/null
+++ b/skills/review/SKILL.md
@@ -0,0 +1,16 @@
+# Review
+
+## Purpose
+Perform a `git diff` from the PR branch to the main branch, and review the code changes for potential issues, improvements, and adherence to best practices. Then provide a plan, pointing to lines of code and modules for refactoring the code to improve its quality, readability, and maintainability.
+
+## Checklist
+- [ ] DO avoid using `hasattr` and `getattr` when the attribute is known to be present.
+- [ ] DO apply `try`/`except` only to the clause that may raise the exception, not to the whole function.
+- [ ] DO pin the exception type in the `except` clause; DON'T use a bare `except`.
+- [ ] DO return or continue early to avoid deep nesting.
+- [ ] DO refactor code with more than three levels of nesting into helper functions.
+- [ ] DO place imports at the top of the file; DON'T place imports inside functions unless there is a specific reason (slow import, circular import, etc.).
+- [ ] DO consider the full set of changes and whether there is a more efficient or cleaner way to achieve the same result.
+- [ ] DO use consistent and readable naming conventions, e.g., if `FollowUpSuggestion` is used, the variable name should be `follow_up_suggestion`, not `followup_suggestion` or `follow_up_suggestions` (plural)
+- [ ] DO ensure the feature or enhancement documented in the docs
+- [ ] DO NOT use implicit boolean checks on pandas objects, e.g. `if df`, `if df == obj`, etc