Trustfull

Score phone numbers, email addresses and IPs for fraud risk from Claude.

Opens Agent Studio, where connecting is one click. The connector URL below works in any MCP client.

Paste it into any MCP client. Setup docs

What the Trustfull connector does

The Trustfull MCP server brings Trustfull's fraud intelligence into Claude. Fraud and risk teams can score phone numbers, email addresses and IP addresses, read the reason codes behind each score, and ask Claude to explain the risk. Trustfull says all five tools are read-only; each scoring call spends one credit on the API key chosen at sign-in.

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.

Trustfull tools (5)

  • get_product_fields_catalog
  • get_reason_codes_catalog
  • score_email
  • score_ip
  • score_phone

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

Limits

  • Scores cost credits. Keep tool approval on, especially with production keys, as Trustfull advises.
  • Calls time out after 30 seconds, per Trustfull.
  • Untrusted input can steer the agent. Trustfull warns about prompt injection in automated pipelines and advises validating inputs first.
  • A score is a signal, not a decision. The connector returns risk data; your team decides what to do with it.
  • No tool schemas from the live server. The tool list needs OAuth; parameters come from Trustfull's docs.
  • We called no tool. Our probe was an anonymous initialize and the RFC 9728 metadata.

Frequently asked questions

Why does Claude not see the Trustfull tools?

Usually the server URL or the client needs a fix. Trustfull's troubleshooting guide says to restart the client after adding the connector, and to check the URL, which its docs give with a trailing slash. It also offers a public health check at mcp.trustfull.com/status to confirm the server answers from your network. Behind a corporate firewall, allow outbound HTTPS to mcp.trustfull.com on port 443.

Why does a Trustfull tool call return an authentication error?

The sign-in or the API key behind it is no longer valid. Trustfull's guide says to sign in again from your client, then check that the API key you picked on the consent screen is still active and has credits in the Trustfull dashboard. The connector has no API key parameter, so a new key only takes effect after you reconnect and select it during sign-in.

Sources

Use it in an agent

Put Trustfull to work.

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

Call (650) 285-1019, Agentman.Our AI receptionist answers.