Skip to content

Attack detection

Attack detection tells you who is probing your store and whether any of it worked. Sill checks every request your Cloudflare Worker or WordPress plugin sees against 17 kinds of attack, groups what it finds by sender, and lists the senders under Security → Attackers, newest first. The Security summary shows the same period at a glance.

Detection reads only the request itself: the method, the address and query string, the user agent and the referring page. It never reads what a visitor submits in a form or a request body.

In the listWhat the sender did
Looked for secret filesAsked for files that hold passwords and keys, such as /.env, /.git/config or /wp-config.php.
Looked for backup filesAsked for backup copies and database dumps, such as /wp-config.php.bak or /site.sql.
Probed debug pagesAsked for pages that reveal server details, such as /phpinfo.php.
Probed a vulnerable pluginAsked for a file from a plugin with a known hole.
Looked for webshellsAsked for a backdoor script a previous attacker might have left, such as /shell.php.
Tried to run a script in uploadsAsked your uploads folder to run a script.
Tried a disguised script fileAsked for a script hidden behind a second extension, such as photo.jpg.php.
Tried path traversalUsed ../ to reach files outside your site.
Tried SQL injectionPut database commands in the address or query.
Tried shell commandsPut server commands in the address or query.
Tried template injectionPut template code in the address or query.
Tried Log4ShellSent the Log4Shell attack string.
Used a credential scannerIdentified itself as a tool built to find leaked keys.
Listed WordPress usersAsked WordPress for its list of user names, a first step towards guessing passwords.
Repeated login attemptsHit your admin login page again and again in a short time.
Scanned for missing pagesAsked for many pages that don’t exist in a burst, looking for something that does.
Posed as a known crawlerUsed a known crawler’s name without being it. See Impostors and scanners.

Open Security → Attackers. The list shows probe campaigns first, then ATTACKS: one card per sender.

Attackers list: two campaigns, then four sender cards from the Netherlands, Germany, the United States and Singapore, one marked succeeded in red One card per sender. Red text means at least one request got a success response.

Each card shows:

  • how serious it is, from the icon (critical, high, medium or low);
  • where the sender is, with a flag, and what they tried, in one sentence, for example Someone from Hesse, Germany tried path traversal.;
  • the address and its network (a hosting company, or Home or mobile network);
  • how many requests, and how long ago;
  • A request succeeded in red when any of their requests got a success response from your site;
  • Blocked with the end date, when you have blocked the sender;
  • NEW next to the address when the sender was first seen on your site in the last 24 hours;
  • seen before · N in 30 days when this address raised alerts here before the current window;
  • also on N other Sill sites when the same address probed other Sill-protected sites in the last 24 hours — this is a count only, never a list of sites.

Above the list:

  • Search address or network narrows the list, for example to 203.0.113 or DigitalOcean.
  • ADDRESSES and NETWORKS switch between one card per sender and one row per network, with how many addresses and requests each network sent. Click a network to list its addresses.
  • GOT THROUGH shows only senders with a request that got a success response.
  • BLOCKED opens your block list.

Click a card to open the sender’s panel.

Sender panel: address 203.0.113.77 in North Holland, one request succeeded, what they tried with each request and its response code, why it was flagged, and history When a request succeeded, the panel says so first, and the request list shows which one.

  • WHO — the address, location, network and the user agent it sent.
  • WHAT THEY TRIED — the summary line, a What those mean → disclosure that explains each attack class in plain language, then each request with the time, the address it asked for and the response your site gave. A 200 next to a secret file means your server returned something: check whether that was the real file or your normal not-found page.
  • Copy for my host — one click copies a plain-text block (address, location, network, first and last seen, each request with its time, method, path and status) you can paste into a hosting support ticket.
  • WHY IT’S FLAGGED — Sill’s confidence and the reasons, such as From a hosting provider’s network.
  • HISTORY — whether the sender visited before, and Other sites: how many other sites using Sill it was seen on in the last 24 hours.

From the panel you can block the sender, trust the address or ignore it for 30 days.

  1. Open any card marked A request succeeded and check the file it names on your server. Remove or protect it.
  2. Turn on Attack paths so requests for those files are refused before they reach your server.
  3. Block senders that keep coming back.