SearchBlueprint fetches pages on behalf of its customers — audits, page checks, free tools. Every fetcher identifies itself truthfully in its user-agent, so you can recognize, allow, or refuse the traffic. If our crawler showed up in your logs, this is the reference page.
The user-agent strings
| String contains | Used for |
|---|---|
SearchBlueprintBot | Site crawling on a customer's behalf, with a contact address in the UA string |
SEOAnalyzer | Audit page fetches |
SEOBot | Page metadata reads |
SearchBlueprint-FreeTool | Single-page checks from the public free tools |
Allow or refuse the traffic
- To allow us: add the strings from the table above to your firewall or CDN allowlist. Site owners auditing their own site want this — the audit is reading your site because you (or your agency) asked it to.
- To refuse us: block the same strings. We don't rotate identities to evade blocks.
How the fetchers behave
All of them:
- Make GET requests only, against public pages — no form submissions, no login attempts.
- Run with bounded concurrency and volume. A crawl covers at most a few hundred pages, once, when a customer runs it. There's no continuous background crawling of your site.
- Retry once with browser-style headers when a bot wall answers — a workaround for over-eager filters, not an attempt to hide. The retry serves the same customer request.
Limits and usage
- Crawls originate from cloud infrastructure, so IP allowlisting is less reliable than user-agent rules — use the UA strings.
- The crawler reads robots-level signals for discovery (robots.txt, sitemaps), but note that audits run at a customer's request on their own site — the customer relationship, not robots.txt, is the authorization.
If something isn't working
You see the UA but didn't run an audit. Someone audited your site — a prospect, a competitor doing research, or an agency evaluating you. The traffic is a bounded read of public pages; blocking it is your right.
You want the contact behind a crawl. The SearchBlueprintBot UA carries a contact address — write to it with the timestamps and we can look into the specific crawl.