← Back to software Offline URL analysis

URL Security Engine

Safely inspect URLs, encoded values, and web-related tokens without connecting to the destination.

Overview

URL Security Engine is an offline command-line analysis tool designed to safely inspect URLs, encoded values, and web-related tokens without connecting to the destination.

Security-related URLs can contain far more information than is immediately visible. Query parameters may conceal destinations using Base64, Base64URL, or percent encoding. Internationalized domain names can contain visually similar characters from different writing systems. URLs can also use misleading user-information syntax, nonstandard ports, nested destinations, and structured tokens such as JSON Web Tokens (JWTs).

URL Security Engine decomposes these structures and presents them in a form intended for security analysis.

Key capabilities

URL structure analysis

The engine parses URLs into their major components:

  • Scheme, hostname, and port
  • Path, query string, and fragment
  • User-information fields
  • Individual query parameters

Query parameters are preserved individually, including duplicate names and empty values.

Encoding analysis

URL Security Engine detects and decodes several common encoding formats:

  • Percent-encoded values
  • Base64
  • Base64URL
  • Base64URL values with omitted padding

Decoded values are classified where possible as URLs, email addresses, IPv4 or IPv6 addresses, JSON, GUIDs, file paths, or plain text. Strict UTF-8 validation helps prevent arbitrary binary data from being presented as meaningful encoded text.

Nested URL discovery

Encoded query parameters may contain additional destination URLs. URL Security Engine identifies these destinations, decodes them, and reports their structure while preserving the parameter and encoding method from which the destination originated.

This makes it easier to determine where an encoded URL actually leads without navigating to it.

IDN and Punycode analysis

Internationalized domain names are inspected in both their Unicode and DNS-compatible ASCII representations. The engine can identify:

  • IDN/Punycode hostnames
  • ASCII/Punycode and Unicode representations
  • Unicode scripts present in the hostname
  • Mixed-script domain names

In interactive mode, Cyrillic characters are highlighted to make visually deceptive Latin/Cyrillic combinations easier to recognize.

JWT analysis

JSON Web Tokens contained in URL parameters can be identified and decoded offline. The engine exposes common JWT properties, including:

  • Algorithm and token type
  • Issuer, subject, and audience
  • Expiration, issued-at, and not-before times
  • Header and payload content
Important: JWT signatures are not cryptographically validated. The engine distinguishes signature presence from signature verification.

Security indicators

Rather than assigning an arbitrary risk score, URL Security Engine reports specific observable conditions:

  • User-information syntax before the actual hostname
  • Nonstandard ports and literal IP-address hosts
  • IDN/Punycode hostnames and mixed Unicode scripts
  • Base64/Base64URL content and encoded nested URLs
  • JSON Web Tokens

The engine distinguishes informational observations from potentially risky URL characteristics without declaring a URL malicious solely because a particular feature is present.

Standalone value analysis

The --value mode allows encoded strings and tokens to be analyzed without constructing a URL:

URLSecurityEngine.exe --value "<value>"

This is useful for values copied from security telemetry, scripts, logs, email headers, authentication events, or investigation tools.

Machine-readable JSON

Analysis can be returned using the stable JSON schema:

URLSecurityEngine.exe "<url>" --json
URLSecurityEngine.exe --value "<value>" --json

JSON output contains no interactive banner or console formatting, making it suitable for scripts, automation, other security applications, and analysis workflows.

Offline by design

URL Security Engine does not navigate to or contact analyzed URLs. Analysis does not require:

  • HTTP requests or DNS lookups
  • Redirect following
  • WHOIS or RDAP queries
  • Reputation services
  • Sandbox execution

URLs are treated strictly as data. This allows potentially hostile URLs to be structurally examined without intentionally contacting the associated infrastructure.