Norton MCP server icon

Norton

by Norton

HIPAA CompliantSOC2 ReadyISO 27001 Ready
Productivity7 tools

Check suspicious messages, links and emails for scams from inside an AI chat, backed by Norton Genie. Anthropic lists 7 tools; Norton publishes no tool list. No Norton account required, and no tool changes any device setting.

Verified connector

Listed by Anthropic as a partner connector in its Connectors Directory.

Connection checked by Agentman on .

Anthropic states this reflects the level of review a connector received, not a security audit.

Connect Norton via MCP

https://ai-connect.norton.com/claude/mcp

Works in any MCP-compatible client. In Agentman Studio it is one click — no config file to edit.

Norton Tools & Capabilities (7)

Tool names from Anthropic's directory listing. This server requires sign-in, so we could not read tool descriptions or parameter schemas.

Limits

  • Norton publishes no tool list for this server. The seven names come from Anthropic's directory. Norton's support article, its launch blog post and Gen Digital's press release describe the connector's capabilities in prose and name no tools, so we could not diff the listing against a vendor enumeration in either direction.
  • We could not verify the endpoint's behaviour at all. Every path on ai-connect.norton.com, including a nonsense control path, returned the same 403 from an AWS load balancer. We read no tool schemas, no safety annotations and no RFC 9728 metadata, and the "no sign-in required" claim is Norton's, not our observation.
  • Norton documents that Genie is not always conclusive. Its Genie support article states Genie will not always be able to determine whether a message is a scam, and the connector's own documentation defines an "unknown" verdict meaning there was not enough data to analyse the item. Norton's guidance for that case is not to interact with the item.
  • A verdict is an assessment, not a guarantee. Norton's editorial note states its products and services may not protect against every type of threat, fraud or crime. A "safe" result should lower your suspicion, not end it.
  • Submitted content is retained and used to improve detection. 30 days for raw submissions, one year for de-identified extracts, per Norton's Products Privacy Notice. Redact before pasting.
  • Four tool names are bare single words. message, email, troubleshooting and recommendations may collide with tools from other connectors in the same agent, depending on whether your client namespaces tool names by server.
  • The recommendation surface is commercial. product_kb_support and recommendations return Norton product information, and Norton's article describes personalised product recommendations. That is a legitimate feature, but it is not a neutral source on which product you need.
  • No rate limits are published. We found no statement from Norton about request limits, quotas, or throttling on this connector, and could not test for them.
  • We found no pricing or metering of any kind. Nothing in the surface spends money or credits, and the connector is free on all Claude tiers per Gen Digital's announcement.

Frequently asked questions

No. None of the seven tools Anthropic lists carries a verb that sets, enables, disables or configures anything, and Norton describes the connector purely as analysis and guidance. Norton Genie assesses a message and tells you what to do next; you take the action yourself. Norton documents no path from this connector to a device, a firewall or a subscription.

Neither. Norton's support article states plainly that the connector does not require a Norton account and works for both paid and free Claude users. Gen Digital's launch announcement adds that Norton is supported across all tiers of Claude. There is no OAuth sign-in step, so no consent screen scopes the connector's access.

No tool in the listing reaches either. Norton Password Manager, Dark Web Monitoring, Privacy Monitor and LifeLock are separate account-bound products, and connecting Norton to Claude involves no account at all. With no sign-in there is no identity for the server to attach a vault or a breach record to.

It leaves Claude and goes to Norton, which keeps it. Norton's Products Privacy Notice states that original text, email or image submissions and any personal information in them, including email headers, are stored for 30 days, and that text extracted from submissions with personal data removed is kept for one year.

Yes, and Norton says so directly. Its Products Privacy Notice states that submissions provided by users help in improving the product's scam detection rate. The Norton Genie support article separately says every use helps Genie get smarter. Treat anything you paste for a scam check as content you have shared with Norton, not as a private lookup.

Seven, according to Anthropic's directory, and there is no second list to check it against. Norton publishes no tool enumeration in its support article, its launch blog post or Gen Digital's press release, none of which names a single tool. The count is therefore Anthropic's alone, and we could not confirm it live.

So the scam check runs without a prompt each time. Norton's support article directs you to set Interactive tools to Always allow so Norton can automatically check for scams. That is a convenience setting, not a security requirement — leaving it at the default simply means Claude asks before each call, which is the safer posture.

Redact it first. Norton documents that the full submission including email headers is retained for 30 days, so a forwarded email carries its sender, recipients and routing metadata to Norton. Norton's support article confirms the connector receives only what you share in that request, but everything you do paste is shared.

Use in Agentman

Connect once and your agents call these tools on their own — on a schedule, in a workflow, with nobody at the keyboard.

Open in Agentman Studio

Server Info

Category
Productivity
Developer
Norton
Tools
7
Domain
ai-connect.norton.com

Using Claude Desktop or another MCP client? Setup docs — the connection URL above works anywhere.