MalCare bot protection is a WordPress security feature that analyzes request behavior to protect sites from automated traffic. Whether it fits your site depends on evidence you should request from MalCare, not on its marketing page.
So, is MalCare legit? It is an established WordPress security product, and the claims below are MalCare's own, attributed to its feature page. Legitimate does not mean proven for your stack. If you run a CDN or WAF in front of WordPress, the questions that matter are where the block happens, which client IP it sees, how it verifies crawlers, and what you can roll back. This article turns each MalCare claim into a request you can put to the vendor. It is a buyer's checklist, not a test result. We have not run MalCare in a lab and do not report one.
What MalCare says its bot protection does
MalCare describes its bot protection as behavior analysis. According to its feature page, it looks at request timing, pages requested and client signals, then compares that activity with what it observes across its network of more than 400,000 sites. MalCare says it identifies and blocks bots across its network, often on their first request.
Three further claims matter to buyers. MalCare says it verifies recognized good bots such as Google and Bing instead of guessing from a bot's name or challenging visitors with a CAPTCHA. It says legitimate integrations and real visitors pass through. And it says AI crawlers are treated like other bots and identified by behavior, with the decision to let them use your content left to the site owner. The page does not describe a separate allow or block control for AI crawlers.
Geoblocking is described as its own owner-controlled feature. You select countries in the dashboard, blocked visitors receive a 403, and the setting can be reversed. MalCare also separates three things: bot protection for automated traffic, geoblocking for chosen countries, and firewall rules for malicious request payloads.
Every statement in this section is MalCare's. Read it as a vendor's description of intent, and read the rest of this article as the list of what to confirm.
What MalCare's claims leave open
A feature page tells you what a vendor wants you to believe. It leaves the operational details open. Several of those details decide whether the product works on a real publisher or ecommerce site.
The network-size figure is a vendor statement. It says nothing about how much of that signal applies to your traffic, or how a decision gets made about your site specifically. The phrase "before WordPress loads" does not say where the check runs: at a hosted edge, at your web server, or inside PHP early in the request. The page describes crawler verification but, in the material we reviewed, gives no method, such as reverse DNS, published IP ranges or signed requests. It says legitimate integrations pass through without saying how you test that for your own webhooks and payment callbacks. And it does not tell you what you can see afterward: a log of blocked requests, the reason for each block, and how to reverse a decision.
None of these gaps is a fault. They are the questions. The next sections turn each into something you can ask for in writing. Don't assume answers for features the page does not state.
Where enforcement happens and which client IP it sees behind a CDN
Ask MalCare one question first: at which hop does the block happen, and what source address does that hop evaluate?
This matters most if a CDN or WAF already sits in front of WordPress. If the check runs after your CDN, the direct peer your origin sees is the CDN, not the visitor. A tool that keys its decisions on that address will either see every request as one client or trust a forwarded header it should not trust. Both outcomes are bad. One blocks real visitors in bulk. The other lets an attacker choose the address by sending a header themselves.
Request these specifics:
- The enforcement location, stated plainly: edge service, web server layer or application layer.
- The header or mechanism used to recover the real client IP behind a proxy, and which proxy addresses it trusts.
- What happens when the forwarded header is missing, malformed or contains several addresses.
- Whether decisions are made on the address alone or on behavior across a session.
You can check the answer yourself. Send a test request through your CDN with a spoofed forwarded address and confirm the log records the address you expect. Our guides to Nginx bot protection and real IPs and Laravel trusted proxies show the same trust-boundary problem in other stacks. The principle carries over to WordPress: trust a forwarded address only from proxies you control.
Verified crawlers, AI crawlers and geoblocking: what policy evidence to ask for
MalCare says it verifies Google and Bing rather than trusting their names. Ask what "verifies" means. The credible methods are reverse and forward DNS confirmation, published IP ranges, or cryptographic signatures where a crawler offers them. A user agent string alone is not verification, because anyone can send it.
This is a practical worry. Centinel's published research found that on a typical news site, 84% of requests claiming to be GPTBot were fake. A product that trusts the name lets those requests through under a trusted label.
For AI crawlers, MalCare's page frames the decision as yours but describes detection by behavior, not a per-crawler policy. Ask for the policy surface in concrete terms:
- Can you allow one AI crawler and block another, and where is that set?
- If an AI crawler is verified, does that verification change the decision, or is it still judged on behavior?
- How are unverified requests that claim a crawler identity handled?
For geoblocking, ask what a blocked visitor sees (MalCare says a 403), and whether the setting is reversible without support. Also plan for the known limit: country rules can miss visitors whose traffic appears to originate from an allowed country, such as through a VPN. Geoblocking is a coarse filter for a known problem. It is not bot detection.
If you also run Cloudflare rules, compare how those handle verified bots, as in Cloudflare bot score thresholds by route, so two layers don't contradict each other.
Test legitimate integrations before you rely on it
MalCare says legitimate integrations pass through. Don't take that on faith. Make a list and test it on staging before enforcement touches production.
Start with everything that calls your site without a browser:
- Payment and order webhooks from your processor.
- Uptime and synthetic monitors.
- Your own mobile app or headless front end calling the REST API.
- Syndication partners and feed readers.
- Search engine crawlers you verified in the previous step.
- Editors and contractors logging in from changing networks or shared offices.
For each, send a real request and record the response code and whether a block was logged. Repeat after any change to your CDN rules, because a new challenge or header rule upstream can change what the plugin sees. The exception patterns in Cloudflare bot exceptions for webhooks and uptime checks apply here too: keep each exception narrow, owned and dated.
For login and XML-RPC specifics, our guide to WordPress login and XML-RPC controls covers endpoint hardening separately. If your host runs its own layer, Cloudways IP exceptions shows how to identify which layer returned a 403 before you change the wrong setting. A failed integration test is a vendor question, not proof of a defect, until you know which layer blocked it.
Logs, false positives and rollback: the evidence pack to request
Before you rely on any bot control, ask for the evidence you would need on your worst day: a legitimate customer or partner reports being blocked and you must explain why.
Request a concrete pack from MalCare and confirm each item exists before buying:
- Per-request log: timestamp, client IP as seen, route, decision and the reason for it.
- Retention and export: how long logs are kept and whether you can export them to your own tooling.
- Decision modes: whether you can run in a log-only or monitor stage before enforcing.
- Override path: how to allow a specific address, route or crawler, and how quickly it takes effect.
- Rollback: how to disable bot protection or geoblocking entirely, without a support ticket, and what state the site returns to.
- False-positive process: how a blocked visitor or partner can be identified and cleared.
Then run a rollout in stages: log, challenge or restrict one route, then enforce, as described in our WordPress guide. Define success before you start. Blocking more sessions is not success if real users can no longer log in or check out. Record both suspicious requests and completed legitimate actions, and compare them before and after.
If MalCare can't show you a reason for a block, you can't defend the decision to your own team. That alone should lower your confidence, whatever the network size.
When a WordPress plugin is not the whole answer
A WordPress security plugin protects WordPress. If your exposure is broader, check whether one layer is enough.
Three situations push past a plugin alone. First, you serve content scrapers want, and they run real browsers behind residential proxies, where address-based and behavior-from-one-site signals are weakest. Second, your traffic includes APIs, checkout or login outside WordPress. Third, you already run a CDN or WAF and need consistent crawler policy across all of it, not a second, separate rule set on one application.
Centinel Analytica is bot and AI crawler protection that tells real visitors from scrapers and applies allow, challenge or block policy through your existing CDN or edge. It works alongside Cloudflare, CloudFront, Akamai and Fastly, so your CDN or WAF stays in place. The two address different layers, and the right question is which gaps your evidence pack above exposes.
Compare your CDN-level options in Cloudflare Bot Management alternatives as well.
Frequently asked questions
The answers below cover the questions buyers ask most about MalCare bot protection, with MalCare's own claims attributed to its feature page.
Ask for the evidence pack, then test it against your own traffic
MalCare bot protection may suit a WordPress site with no other controls, but its feature page is a claim, not proof for your stack. Get the enforcement location, the client IP method, the crawler verification basis, an integration test result and a rollback path in writing before you rely on it.
Frequently asked questions
Is MalCare legit?
MalCare is an established WordPress security product, and its bot protection claims are described on its own feature page. Legitimate is not the same as proven for your setup. Ask MalCare for the enforcement location, how it recovers client IPs behind a CDN, how it verifies crawlers, and what logs and rollback you get.
Does MalCare bot protection work behind Cloudflare or another CDN?
MalCare's feature page does not state how it behaves behind a CDN, so ask. The key question is which client IP it evaluates when your CDN is the direct peer, and which proxies it trusts. Test with a spoofed forwarded address and check the logs. Centinel Analytica runs alongside Cloudflare, CloudFront, Akamai and Fastly, leaving your CDN in place.
Does MalCare block AI crawlers like GPTBot?
MalCare says AI crawlers are treated like other bots and identified by behavior, with the decision to allow them left to the owner. Its page does not describe a separate allow or block control. Ask how it verifies a crawler's identity, since unverified claims are common: Centinel found 84% of GPTBot-claiming requests fake on a typical news site.
What should I test before enabling MalCare bot protection on production?
Test every non-browser caller on staging: payment webhooks, uptime monitors, your app or REST clients, syndication partners and verified search crawlers. Record the response code and whether a block was logged for each, and repeat after CDN rule changes. Run log-only first if available.
