diff --git a/.serena/memories/clojure/idioms.md b/.serena/memories/clojure/idioms.md
new file mode 100644
index 0000000000..c8672aa647
--- /dev/null
+++ b/.serena/memories/clojure/idioms.md
@@ -0,0 +1,10 @@
+# Clojure Idioms (verified)
+
+Behaviors confirmed against the language/stdlib — do not re-derive from
+assumption; a wrong assumption here already cost a review round.
+
+- `int?` is NOT 32-bit-only: true for `Long`, `Integer`, `Short`,
+  `Byte` (fixed-precision integers). Clojure integer literals are
+  `Long`, so `(int? 5000)` is true.
+- `integer?` is the general integer predicate; prefer it when any
+  integer kind must match, `int?` only when fixed precision is meant.
diff --git a/.serena/memories/critical-info.md b/.serena/memories/critical-info.md
index d36edd4cfa..892e5687f2 100644
--- a/.serena/memories/critical-info.md
+++ b/.serena/memories/critical-info.md
@@ -7,6 +7,9 @@ You are working on the GitHub project `penpot/penpot`, a monorepo.
    before focused memories.
 - Edits/stale refs/duplication cleanup: `mem:memory-maintenance`.
 - Cross-cutting testing principles, TDD workflow, and anti-patterns: `mem:testing`.
+- Verified Clojure language behaviors that contradict common assumptions
+  (e.g. `int?` covers `Long`; `integer?` is the general predicate):
+  `mem:clojure/idioms` — read before assuming stdlib predicate semantics.
 
 # Development workflow