Resources · Claude connector

Claude McAfee connectorWhat Claude can do in your McAfee account.

The Claude McAfee connector exposes 3 tools, all 3 of them read tools. 0 write, 0 undocumented. Claude sends a dodgy link, email or text to McAfee's analysis and walks you through the verdict. Below: what gets checked and the language rule to know.

Verified Trustpilot reviews · AI, automation & growth agency

Overview

A second opinion before you click

A text says your parcel is stuck, an email demands an urgent payment, a link looks off. With McAfee connected, you paste the thing into Claude and it gets checked against McAfee's threat intelligence. You get a plain verdict, the reasoning behind it, and pointers on what to do next.

Sort out a shady text. check_text_for_scam screens it for phishing, social engineering and suspicious links, and returns an explained verdict.

Judge an email before replying. An invoice lands from a vendor nobody recognizes. check_email_for_scam reviews the content and returns a call, with the reason for it.

Vet a single link. check_website_url looks at the site's reputation (malware, phishing, scams) before you click.

Each tool returns a full assessment: a clear verdict, a human-readable confirmation, how the verdict was reached, and further protection guidance. The text tool covers SMS, RCS, iMessage, WhatsApp and other messengers.

What the connector won't do: McAfee explicitly rules out using the email tool to draft, send or manage an email, and keeps it to safety analysis. By default its sheet only exposes reading. It doesn't watch your inbox either, so nothing happens until you ask. Hands-off workflows that run on their own schedule are a different job, gathered on the Integrations hub.

Vocabulary

Words to know before connecting McAfee

Five terms, explained once.

Connector
The bridge you set up once between Claude and a service you already use, so Claude can work with it while answering you.
Tool
A named capability the connector opens to Claude. Claude picks what it needs on its own; the directory sheet lists every one.
Authorization
The service's own screen where you hand Claude the access it will use. Granted once per person, and revocable.
Approval
The confirmation Claude waits for before an action that changes something in an account, shown in the chat at the right moment.
MCP
The open standard connectors are built on: it lets an assistant like Claude hold a conversation with an outside service.
Connect

Connect the Claude McAfee connector

  1. 01

    Find McAfee in Claude

    In Claude, open Customize, then Connectors, and spot McAfee. On a Team or Enterprise workspace, an Owner or Primary Owner enables it for the organization first; each member then connects on their own.

  2. 02

    Start the connection

    Click Connect on the McAfee row, then follow the window the service opens. If the link ever breaks, hit Disconnect and connect again from the same row.

  3. 03

    Read the authorization screen

    Go through the authorization screen before you accept. It belongs to the service, not to Claude, and it sets what the access covers. You can also take that access back later from the service itself.

Tools

McAfee's 3 analysis tools

McAfee gives Claude 3 tools: 3 that read your account, 0 that change something in it.

Three checking tools, described by the tool list McAfee's service publishes. Names stay in English, the way Claude displays them.

  • 3 read

What Claude checks (3)

3 tools

Three tools that analyze what you share without changing anything.

check_email_for_scam

Has McAfee inspect an email's content for phishing, social engineering or other scam markers. The result states whether it's a scam, why, and what you can do next.

When it helps
an email asks you to pay into new bank details and you want a read on it before acting.
Watch out
McAfee scopes it to safety analysis, not to managing your mailbox.

Sourceclaudecompanion.gateway.api.mcafee.com · October 1, 2026 ↗

check_text_for_scam

Submits a short message (SMS, RCS, iMessage, WhatsApp or another messenger) for McAfee's analysis. Claude relays the verdict, the explanation, and protective advice.

When it helps
a so-called courier asks for customs fees over WhatsApp, payment link included.
Watch out
paste the text in its original language, since McAfee turns away translated content.

Sourceclaudecompanion.gateway.api.mcafee.com · October 1, 2026 ↗

check_website_url

Rates a web address against McAfee's threat intelligence, looking for malware, phishing or scam signals. You get a safe or not-safe call, explained, with further avenues to protect yourself.

When it helps
you're unsure about a link from an unknown sender.

Sourceclaudecompanion.gateway.api.mcafee.com · October 1, 2026 ↗

Approvals

What Claude asks first, and who can switch McAfee off

By default, Claude stops and asks for your go-ahead before each action it takes on an account for you. The three McAfee tools only analyze what you share, and no source describes a confirmation specific to any of them.

On Team and Enterprise, workspace owners decide whether a member may let certain actions through without being asked again. They can also limit what McAfee may do for the whole organization, keeping reading open and closing writes for example, with a setting nobody bypasses from their own account. Finally, Claude works with the connected person's rights: what that person can't reach, Claude can't reach either.

Plans

Which plans include it

None of the 819 directory sheets shows plan availability. Connector by connector, that information isn't published anywhere: it's a real gap in the catalogue.

The general rule is public: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile, while desktop extensions install on Claude Desktop. On Team and Enterprise, an Owner or Primary Owner opens the connector for the organization first. For the current status on your account, the only reference is the McAfee sheet in the official directory, which the button on this page opens.

Limits

Where McAfee stops inside Claude

A connector isn't an automation: Claude calls these tools while answering you, and none of them watches incoming messages. Only McAfee, the publisher, documents this connector, and the help article linked from its sheet couldn't be read.

The sheet exposes reading only, which is an observed floor rather than a guarantee: an admin can open more than any public page shows. The directory badge isn't a security audit, and Anthropic notes on every sheet that it doesn't control a publisher's tools or how they behave. A connected community connector has exactly the same access as any other. So stick to publishers you know. Other connectors are gathered on the Claude connectors page.

Need help

Need help connecting McAfee to Claude?

A person reads every message.

FAQ

Claude McAfee connector: common questions

01What can Claude do with the McAfee connector?
Claude can have McAfee analyze a link, an email or a short message to tell whether it's a scam. The three tools on the sheet cover those three cases, and each returns a verdict, an explanation of the call and tips to stay safe. You paste the suspicious content into the chat, Claude picks the right tool and relays McAfee's answer in plain words. The connector sticks to security and scam questions, nothing else.
02Can Claude send or delete messages through McAfee?
No, based on the available sources: all three tools are analysis tools, and McAfee states they aren't for drafting, sending or managing an email. No source describes a tool that writes or erases anything, and the service itself flags all three of its tools as read-only. Still, a list that shows only reading is an observed floor, not a guarantee, because an admin can open capabilities missing from the public sheets.
03Does Claude ask before using McAfee?
The default rule is that Claude asks for confirmation before each action it takes on an account for you. No source describes a dedicated confirmation for check_email_for_scam, check_text_for_scam or check_website_url, so nothing McAfee-specific is announced. On a Team or Enterprise workspace, owners decide whether members can let certain actions through without a prompt, and they can restrict the connector for the whole organization at once, with no exception per member.
04Which Claude plans support the McAfee connector?
No official source publishes plan availability connector by connector, and McAfee's sheet doesn't show it. The general rule says remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile. On Team and Enterprise, an Owner or Primary Owner has to enable the connector for the organization first. For your own account, the official directory sheet is the reference, since it shows the current status, updated by Anthropic itself.
05Does Claude send my whole conversation to McAfee?
The sources read don't say. What McAfee's service describes is that each tool receives the content to analyze: an email, a message or a web address. The support article linked from the sheet, which might spell out what McAfee keeps, couldn't be read. On Claude's side the general rule holds: it works with the connected person's rights and nothing more. When in doubt, paste only the suspicious message, not the rest of the thread.
06Why does McAfee reject a translated message?
Because the service detects the content's language on its own and turns down any call where the text was translated or doesn't match the declared language. The language parameter has to reflect the language of the original message. If the language isn't supported, Claude has to relay that refusal as its final answer. No source we read lists the supported languages, since the support article couldn't be read. So always paste the message in its original language.
07Claude or an automation tool for McAfee?
They do different jobs, so it depends on the use. Claude with McAfee handles a one-off doubt about a message: you paste a message, it gets analyzed, and Claude explains the verdict in the chat. An automation tool is meant to chain steps that run unattended, with no conversation. This connector watches no inbox and never starts by itself; it waits for you to ask the question first, every single time.