diff --git a/README.md b/README.md
new file mode 100644
index 0000000..5e5b81e
--- /dev/null
+++ b/README.md
@@ -0,0 +1,31 @@
+# LXX Morphology Downloader
+
+This repository contains a small utility for downloading the LXX Morphology corpus from the UPenn gopher mirror.
+
+## Usage
+
+```bash
+python scripts/download_lxxmorph.py --output-dir lxxmorph
+```
+
+By default, it pulls from:
+
+```
+https://ccat.sas.upenn.edu/gopher/text/religion/biblical/lxxmorph/
+```
+
+The downloader defaults to fetching only `*.mlxx` files so the local corpus stays limited to the MLXX data.
+You can override the base URL with `--base-url` or adjust the filename filter with `--include` if the mirror
+changes or you need additional file types.
+
+## Chat question template
+
+Use the template below when you want an analysis run based on the local `*.mlxx` files and the
+`MT-LXX_Parallel_Manual (1).pdf` instructions:
+
+```
+Please analyze <BOOK CHAPTER:VERSE> using the local .mlxx files. Follow the MT-LXX_Parallel_Manual (1).pdf
+step-by-step and return rows in this format:
+
+Reference,Hebrew (MT),Greek (LXX),CATSS Notation & Analysis
+```
diff --git a/scripts/download_lxxmorph.py b/scripts/download_lxxmorph.py
new file mode 100644
index 0000000..cb2716c
--- /dev/null
+++ b/scripts/download_lxxmorph.py
@@ -0,0 +1,126 @@
+#!/usr/bin/env python3
+"""Download the LXX Morphology corpus directory from UPenn's gopher mirror."""
+from __future__ import annotations
+
+import argparse
+import html.parser
+import os
+import pathlib
+import re
+import urllib.parse
+import urllib.request
+
+DEFAULT_BASE_URL = "https://ccat.sas.upenn.edu/gopher/text/religion/biblical/lxxmorph/"
+
+
+class LinkParser(html.parser.HTMLParser):
+    def __init__(self) -> None:
+        super().__init__()
+        self.links: list[str] = []
+
+    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
+        if tag.lower() != "a":
+            return
+        for key, value in attrs:
+            if key.lower() == "href" and value:
+                self.links.append(value)
+
+
+def fetch_bytes(url: str, user_agent: str | None = None) -> bytes:
+    request = urllib.request.Request(url)
+    if user_agent:
+        request.add_header("User-Agent", user_agent)
+    with urllib.request.urlopen(request) as response:
+        return response.read()
+
+
+def parse_links(content: bytes, base_url: str) -> list[str]:
+    text = content.decode("utf-8", errors="replace")
+    if "<a" in text.lower():
+        parser = LinkParser()
+        parser.feed(text)
+        return [urllib.parse.urljoin(base_url, link) for link in parser.links]
+
+    links: list[str] = []
+    for line in text.splitlines():
+        line = line.strip()
+        if not line or line.startswith("#"):
+            continue
+        if line.startswith("/"):
+            links.append(urllib.parse.urljoin(base_url, line))
+    return links
+
+
+def ensure_directory(path: pathlib.Path) -> None:
+    path.mkdir(parents=True, exist_ok=True)
+
+
+def save_file(url: str, output_dir: pathlib.Path, allow_pattern: re.Pattern[str] | None, user_agent: str | None) -> None:
+    parsed = urllib.parse.urlparse(url)
+    filename = os.path.basename(parsed.path)
+    if not filename:
+        return
+    if allow_pattern and not allow_pattern.search(filename):
+        return
+    target_path = output_dir / filename
+    if target_path.exists():
+        return
+    data = fetch_bytes(url, user_agent=user_agent)
+    target_path.write_bytes(data)
+
+
+def download_recursive(
+    base_url: str,
+    output_dir: pathlib.Path,
+    visited: set[str],
+    allow_pattern: re.Pattern[str] | None,
+    user_agent: str | None,
+) -> None:
+    if base_url in visited:
+        return
+    visited.add(base_url)
+
+    content = fetch_bytes(base_url, user_agent=user_agent)
+    links = parse_links(content, base_url)
+    for link in links:
+        if link.endswith("/"):
+            subdir_name = os.path.basename(os.path.normpath(urllib.parse.urlparse(link).path))
+            subdir_path = output_dir / subdir_name
+            ensure_directory(subdir_path)
+            download_recursive(link, subdir_path, visited, allow_pattern, user_agent)
+        else:
+            save_file(link, output_dir, allow_pattern, user_agent)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Download files from the LXX Morphology directory.")
+    parser.add_argument(
+        "--base-url",
+        default=DEFAULT_BASE_URL,
+        help="Base URL for the LXX Morphology directory listing.",
+    )
+    parser.add_argument(
+        "--include",
+        default="\\.mlxx$",
+        help="Regex pattern of filenames to download (default: only .mlxx files).",
+    )
+    parser.add_argument(
+        "--output-dir",
+        default="lxxmorph",
+        help="Directory to store downloaded files.",
+    )
+    parser.add_argument(
+        "--user-agent",
+        default="Mozilla/5.0 (compatible; LXXMorphDownloader/1.0)",
+        help="User-Agent header to send with requests.",
+    )
+    args = parser.parse_args()
+
+    output_dir = pathlib.Path(args.output_dir)
+    ensure_directory(output_dir)
+    allow_pattern = re.compile(args.include) if args.include else None
+    download_recursive(args.base_url, output_dir, set(), allow_pattern, args.user_agent)
+
+
+if __name__ == "__main__":
+    main()