Siggmund
Log inGet started
← Back to the help centre

SiggmundAgent, Siggmund's crawler

SiggmundAgent is Siggmund's web crawler. If you see it in your server logs, this page tells you what it does on your site.

It was called SiggmundBot until October 2026. The name changes, not what it does or how to block it: the old name is still honoured (see "How to block it"). This page keeps its original address, siggmund.com/fr/aide/siggmundbot for the French version.

How to recognise it

Its requests carry this User-Agent header:

text
Mozilla/5.0 (compatible; SiggmundAgent/1.0; +siggmund.com)

The token to look for is SiggmundAgent. Its previous version announced itself as SiggmundBot/1.0: if your firewall or filtering rules target that name, they no longer recognise it. Target the word Siggmund instead, which appears in both.

Why it reads pages

Siggmund measures what AI assistants (ChatGPT, Claude, Perplexity…) say about a brand: whether they cite it, whether they recommend it, and which sources they rely on. Measuring this requires keeping a dated copy of the pages involved, the way one keeps a record.

So SiggmundAgent reads and archives, on behalf of Siggmund's clients:

  • the public pages of those clients' websites;
  • the public pages that AI assistants cite when asked about those brands: articles, comparisons, competitors' websites.

What it does, and what it does not do

  • It reads public pages, with GET requests. It never logs into an account and never fills in a form.
  • It may open a page in a headless browser, to see the content that the page's JavaScript displays.
  • It changes nothing on the sites it reads.

How often

  • It spaces out its requests to any one site: each of its servers sends that site at most one batch of requests per second, and a batch holds ten pages at most.
  • As a rule, a page it has already read is only read again once its copy is out of date: after one hour to one week, depending on the use.

How to block it

SiggmundAgent respects your robots.txt file. To keep it off your whole site:

text
User-agent: SiggmundAgent
Disallow: /
  • The old name is still honoured: a User-agent: SiggmundBot group (for example a Disallow: / written before October 2026) still applies to it. If you have both groups, their rules add up.
  • A group in its name (SiggmundAgent or SiggmundBot) replaces the User-agent: * group for it. Without a group in its name, it follows the * group.
  • It reads your robots.txt again at most once an hour: a change takes effect within the hour.
  • The Crawl-delay directive is not read.
  • One exception: if your site belongs to a Siggmund client, SiggmundAgent keeps reading it, since that client asked us to measure it.

Contact us

A question, unusual behaviour, a request about SiggmundAgent: write to contact@siggmund.com, ideally with an excerpt of your logs (date, requested address, User-Agent).

Last updated: · First published: