diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000..faa0f4b
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1,6 @@
+# Tell GitHub Linguist to ignore generated/example HTML files when calculating
+# the repo's primary language. Without this, large graph.html artifacts in
+# worked/ dominate the byte count and the repo shows as HTML instead of Python.
+worked/**/*.html linguist-vendored=true
+graphify-out/**/*.html linguist-vendored=true
+*.html linguist-detectable=false