diff --git a/.gitignore b/.gitignore
index 6529ac36c..e50b964da 100644
--- a/.gitignore
+++ b/.gitignore
@@ -17,6 +17,10 @@ plugins/*/compiled
 .cache-loader
 static/llms.txt
 
+# Files generated by script validate_queries.py
+all_queries.sql
+failed_queries.sql
+
 # Misc
 .DS_Store
 .env.local
diff --git a/scripts/validate_queries.py b/scripts/validate_queries.py
new file mode 100644
index 000000000..d497ae45e
--- /dev/null
+++ b/scripts/validate_queries.py
@@ -0,0 +1,123 @@
+import re
+import requests
+from pathlib import Path
+import sys
+import json
+import argparse
+
+parser = argparse.ArgumentParser(
+    description="Execute all QuestDB demo SQL queries found in Markdown files."
+)
+parser.add_argument(
+    "--path",
+    default=".",
+    help="Root folder to search for .md and .mdx files (default: current directory)",
+)
+parser.add_argument(
+    "--url",
+    default="http://localhost:9000",
+    help="QuestDB REST API base URL, e.g. http://localhost:9000 (default: localhost)",
+)
+args = parser.parse_args()
+
+
+QUESTDB_REST_URL = f"{args.url.rstrip('/')}/exec"
+ROOT_DIR = args.path
+ALL_FILE = Path("all_queries.sql")
+FAILED_FILE = Path("failed_queries.sql")
+TIMEOUT = (3, 5)
+MAX_BYTES = 8192
+
+block_re = re.compile(
+    r"```questdb-sql[^\n]*title=\"([^\"]+)\"[^\n]*\bdemo\b[^\n]*\n(.*?)```",
+    re.DOTALL,
+)
+
+def extract_blocks(root_dir):
+    root = Path(root_dir)
+    for ext in ("*.md", "*.mdx"):
+        for path in root.rglob(ext):
+            text = path.read_text(encoding="utf-8", errors="ignore")
+            for match in block_re.finditer(text):
+                title, sql = match.groups()
+                yield path, title.strip(), sql.strip()
+
+def execute_query(query):
+    """Execute the query via GET, keeping it exactly as-is and parsing 400 bodies."""
+    try:
+        with requests.get(
+            QUESTDB_REST_URL, params={"query": query}, timeout=TIMEOUT, stream=True
+        ) as r:
+            content = b""
+            for chunk in r.iter_content(chunk_size=1024):
+                content += chunk
+                if len(content) > MAX_BYTES:
+                    break
+
+            text = content.decode("utf-8", errors="ignore").strip()
+            if r.status_code != 200:
+                try:
+                    js = json.loads(text)
+                    if "error" in js:
+                        return False, js["error"]
+                except Exception:
+                    pass
+                return False, f"HTTP {r.status_code} {r.reason}: {text or 'No body'}"
+
+            if not text:
+                return True, None
+
+            try:
+                js = json.loads(text)
+                if "error" in js:
+                    return False, js["error"]
+                return True, None
+            except json.JSONDecodeError:
+                return True, None
+
+    except requests.Timeout:
+        return False, f"Timeout after {TIMEOUT[1]}s"
+    except Exception as e:
+        return False, str(e)
+
+if __name__ == "__main__":
+    success = 0
+    failed = 0
+    failed_list = []
+    blocks = list(extract_blocks(ROOT_DIR))
+    total = len(blocks)
+
+    print(f"QuestDB REST URL: {QUESTDB_REST_URL}")
+    print(f"Searching in: {Path(ROOT_DIR).resolve()}")
+    print(f"Found {total} queries to execute.\n")
+
+    with ALL_FILE.open("w", encoding="utf-8") as all_out, FAILED_FILE.open("w", encoding="utf-8") as fail_out:
+        for i, (file_path, title, sql) in enumerate(blocks, 1):
+            print(f"[{i}/{total}] Executing: {file_path}  [{title}]")
+            sys.stdout.flush()
+
+            ok, err = execute_query(sql)
+            all_out.write(f"-- {file_path}\n--- {title}\n{sql}\n\n")
+
+            if ok:
+                success += 1
+            else:
+                failed += 1
+                failed_list.append((file_path, title, err))
+                print(f"   ❌ Failed: {err}")
+                fail_out.write(f"-- {file_path}\n--- {title}\n{sql}\n-- ERROR: {err}\n\n")
+
+    print("\n============================")
+    print(f"Executed {total} queries")
+    print(f"✅  Succeeded: {success}")
+    print(f"❌  Failed:    {failed}")
+    print("============================")
+
+    if failed_list:
+        print("\n❌ Failed queries summary:")
+        for path, title, err in failed_list:
+            print(f"  - {path}  [{title}]: {err}")
+
+    print("\nResults written to:")
+    print(f"  • {ALL_FILE}")
+    print(f"  • {FAILED_FILE}")