pwneddesal · April 20, 2021 07:56 · Apr 20, 2021
diff --git a/waybackurls.py b/waybackurls.py
@@ -0,0 +1,40 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+import requests
+import sys
+import json
+
+
+def waybackurls(host, with_subs):
+    if with_subs:
+        url = 'http://web.archive.org/cdx/search/cdx?url=*.%s/*&output=json&fl=original&collapse=urlkey' % host
+    else:
+        url = 'http://web.archive.org/cdx/search/cdx?url=%s/*&output=json&fl=original&collapse=urlkey' % host
+    r = requests.get(url)
+    results = r.json()
+    return results[2:]
+
+
+if __name__ == '__main__':
+    argc = len(sys.argv)
+    if argc < 2:
+        print('Usage:\n\tpython3 waybackurls.py <url> <include_subdomains:optional>')
+        sys.exit()
+
+    host = sys.argv[1]
+    with_subs = False
+    if argc > 3:
+        with_subs = True
+
+    urls = waybackurls(host, with_subs)
+    string_urls = ""
+    for i in range(len(urls)):
+        string_urls = string_urls+urls[i][0]+"\n"
+    json_urls = json.dumps(urls)
+    if urls:
+        filename = '%s-waybackurls.json' % host
+        with open('output/waybackurls/'+filename, 'w') as f:
+            f.write(string_urls)
+        print('[*] Saved results to %s' % filename)
+    else:
+        print('[-] Found nothing')