Upload tokenizer.json

tokenizer.json

@@ -0,0 +1,48 @@
+{
+  "model": {
+    "type": "BPE",
+    "vocab": {
+      "!": 0,
+      "\"": 1,
+      "#": 1
+    }
+  },
+  "description": "Benign sample data: looks like phishing or payloads in places, but is safe when treated as JSON only.",
+  "clickable_url": {
+    "label": "Example documentation link",
+    "href": "https://example.com/",
+    "note": "IANA-reserved domain; safe placeholder for link tests."
+  },
+  "embedded_image": {
+    "alt": "Small SVG inline as data URL (green square only)",
+    "src": "data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='16' height='16'%3E%3Crect fill='%23008000' width='16' height='16'/%3E%3C/svg%3E",
+    "remote_fallback": "https://www.w3.org/Assets/w3c_logo.svg"
+  },
+  "looks_like_credentials_but_is_fake": {
+    "username": "demo_user_not_real",
+    "password": "hunter2_is_a_meme_not_a_secret"
+  },
+  "escaped_html_as_data_not_dom": {
+    "html_fragment": "<a href=\"https://example.org/\">Click me (string only)</a><img src=\"https://www.w3.org/Assets/w3c_logo.svg\" alt=\"W3C\" />"
+  },
+  "javascript_url_as_literal_string": {
+    "do_not_use_as_href": "javascript:alert('This is only text inside JSON; it does not execute here.')"
+  },
+  "shell_like_strings": {
+    "command_looking": "curl -s https://example.com/ | head -n 1",
+    "note": "Plain string; not executed by JSON parsers."
+  },
+  "base64_looking_but_harmless": {
+    "payload": "SGVsbG8sIHRoaXMgaXMganVzdCBiYXNlNjQgZW5jb2RlZCB0ZXh0Lg==",
+    "decoded_hint": "Decodes to a simple English sentence, not binary malware."
+  },
+  "ipv4_that_looks_suspicious": {
+    "address": "127.0.0.1",
+    "context": "Loopback; common in examples, not an attack by itself."
+  },
+  "unicode_homoglyph_example": {
+    "display": "exаmple.com",
+    "warning": "Contains Cyrillic 'а' (U+0430) instead of Latin 'a' — safe to store, but teaches URL review.",
+    "safe_ascii_equivalent": "example.com"
+  }
+}