diff --git a/README.md b/README.md
index 9909d26..7ed8083 100644
--- a/README.md
+++ b/README.md
@@ -23,11 +23,11 @@ programmatic/LLM consumption.
 
 ```mermaid
 flowchart TD
-    A["tokenize(text)"] --> B["parse_sexp(text)\n{ tag, items[] } tree"]
-    B --> C["run_query(ext, query, root, args)\nfile-type + query dispatch"]
+    A["tokenize(text)"] --> B["parse_sexp(text)<br>{ tag, items[] } tree"]
+    B --> C["run_query(ext, query, root, args)<br>file-type + query dispatch"]
     C --> D["plain data object"]
-    D --> E["render_human(ext, query, data)\nformatted tables for terminal"]
-    D --> F["render_json(data)\nJSON.stringify for LLM / scripts"]
+    D --> E["render_human(ext, query, data)<br>formatted tables for terminal"]
+    D --> F["render_json(data)<br>JSON.stringify for LLM / scripts"]
 ```
 
 Query functions are pure data transformers — no I/O, no format awareness.