diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000..ff268c8
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1,7 @@
+# Mark generated folders as generated to exclude from language statistics
+articles/** linguist-generated=true
+html/** linguist-generated=true
+
+# Mark XML files as generated since they're also auto-generated
+articles_all.xml linguist-generated=true
+articles_latest.xml linguist-generated=true