Security Scanners UAs
The user-agent strings that security scanners, port scanners, directory brute-forcers and internet-wide reconnaissance tools send when they probe your servers, ready to copy for log analysis and firewall or WAF rules. The list is updated as new scanners appear.
| Kind of tool | Names found in the user agents |
|---|---|
| Internet-wide scanners and search engines | l9scan, l9explore, l9tcpid, CensysInspect, Cortex-Xpanse (Palo Alto Networks), ICI-probe, Yokohama Institute of Information Security |
| Port and banner scanners | masscan, zgrab, HTTP Banner Detection, goscan-mini |
| Directory and path brute-forcers | gobuster, feroxbuster, pathscan |
| Vulnerability, CMS and exploit scanners | WPScan, WP-Safe-Scanner, EnvScanner, php2shell-scan, MIKASA-scan, TLM-Audit-Scanner, MerchantSecurityScanner, CVE-… tags (e.g. CVE-2026-74759, cve-2026-63030) |
| Bare or generic strings | Mozilla/4.0, Mozilla/5.0, Mozilla/6.0, Mozilla/5.0 (Windows NT 10.0; Win64; x64) |
About this tool
Security scanners, port scanners and reconnaissance tools announce themselves in the User-Agent header of their HTTP requests, at least the ones that do not try to hide: a project name and version (masscan, zgrab, gobuster, WPScan), a link to the organisation running an internet-wide scan (LeakIX, Censys, Palo Alto Networks Cortex Xpanse, academic research), a CVE identifier when a tool probes for one vulnerability, or simply a bare "Mozilla/5.0" left at its default. These tools map attack surfaces: which hosts, ports, services, software versions and paths are exposed, and internet-wide scanners may publish the results in searchable databases. This reference gathers the user-agent strings seen in real server logs in one copy-ready list, with a table that groups them by what the tool does, so you can recognise scanning and reconnaissance traffic in access logs, analytics and security dashboards.
Such traffic is not stopped by robots.txt, which scanners rarely respect; block or rate-limit it with Nginx or Apache rules, in your WAF or at the firewall, and verify the source with the Reverse DNS Lookup and the IP Blacklist Check before deciding, since legitimate audits and research scans use the same tools.
Related: AI Crawlers and Bots, My User Agent, Port Scan, HTTP Headers.
How to use it — quick steps
- 01
Copy the list
Press [copy] on the box and paste the strings into your log filter, SIEM query, WAF rule or bot-management tool.
- 02
Match by name
For reliable matching use the tool name from the table (masscan, zgrab, gobuster, l9scan…) rather than the whole string, since versions change.
- 03
Act on behaviour and source
Combine the user agent with the request pattern and the requesting IP before blocking; a bare "Mozilla/5.0" alone is only a hint.
Key features of this tool
Copy-ready list
One user-agent string per line, exactly as the tools send it, with a [copy] link; no wrapping, so each line is a complete string.
Grouped by purpose
A table that sorts the strings into internet-wide scanners, port and banner scanners, brute-forcers, vulnerability scanners and bare generic strings.
Kept up to date
New scanners and new versions of existing ones are added as they show up in server logs.
Frequently asked questions
What are these tools doing on my server?
Mapping the attack surface: which hosts answer, which ports and services are open, which software and versions run, which paths exist (admin panels, backups, .env files) and which known vulnerabilities apply. Internet-wide scanners such as LeakIX, Censys or Cortex Xpanse publish what they find in searchable databases; brute-forcers and CVE probes are usually the first step of an attack or of a penetration test.
Can I stop them with robots.txt?
No. robots.txt is a request that well-behaved search engines honour; scanners and attack tools ignore it in most cases. Block or rate-limit them in Nginx or Apache rules, in your WAF or at the firewall, by user agent and, better, by the behaviour of the requests.
Why are "Mozilla/5.0" and "Mozilla/4.0" in the list?
Real browsers never send such a bare string; it is what many scanners and scripts use by default. Treat it as a signal together with the request pattern (many 404s, probing of /wp-login.php, /.env, /.git), not as proof on its own.
Is every request with these user agents hostile?
Not necessarily: research projects, your own security team's audits and services such as a payment provider's merchant scanner use them legitimately, and a user agent can be spoofed by anyone. Decide on the source IP and the behaviour, not on the string alone; the Reverse DNS Lookup and the IP Blacklist Check help identify the sender.