MCP

Connect your AI agents (Cursor, VS Code, Claude, ChatGPT) directly to Delpha's Data Quality APIs using the open Model Context Protocol (MCP).

Model Context Protocol ( MCP )

Let your AI agents call the Delpha Data Quality APIs over the Model Context Protocol.

The Delpha MCP server exposes the same Data Quality tools as the Delpha API — Find, Validate, Enrich, Score — through the open MCP standard, so any MCP-compatible client (Cursor, VS Code, Claude Code, Claude.ai, ChatGPT) can use them directly. Authentication is OAuth-only, scoped per tenant.

Server URL — https://mcp.delpha.ai/mcp

Auth — OAuth 2.1 with PKCE (no API key needed). Your Delpha admin provisions your account; you finish activation from the Cognito invite email and then connect from your MCP client.

Connect to Delpha's MCP server

Cursor

Install in Cursor

To open Cursor and automatically add the Delpha MCP, click install. Alternatively, add the following to your ~/.cursor/mcp.json file. To learn more, see the Cursor documentation.

{
  "mcpServers": {
    "delpha": {
      "url": "https://mcp.delpha.ai/mcp"
    }
  }
}

VS Code

Claude Code

Claude.ai (web)

ChatGPT

Other MCP Clients

Authentication

Delpha MCP uses OAuth 2.1 with PKCE. There is no API-key fallback — every call is bound to a Delpha user account so usage stays auditable per tenant.

How it works:

1. Your Delpha admin onboards your user (an invite email is sent automatically). The invite carries a one-time activation code and a link to the Delpha Cognito hosted UI.

2. You activate the account, set a password, and (optionally) enroll MFA.

3. From your MCP client, add the connector with the server URL above. The client opens the Cognito hosted UI for consent.

4. Once consented, the client receives an access token bound to your tenant. The MCP server validates the token on every request, looks up your tenant's API credentials, and routes the call to the Delpha API.

Need an account? Ask your Delpha administrator — onboarding is admin-driven for compliance reasons. Contact [e-mail] if you don't know who that is.

Tools

The MCP server exposes 14 tools, grouped by workflow below. Every tool maps to a Delpha API endpoint — click API reference on any row to see the full request/response schema on the existing API page.

Find & Validate

MCP tool	Objective	Description	Live lookup	API reference
findAndValidateEmail	Recommend / Validate	Find & validate email	Yes	email
findAndValidateLegalID	Recommend / Validate	Find & validate legal ID	Yes	legal-id
findAndValidateLinkedin	Recommend / Validate	Find & validate LinkedIn URL	Yes	linkedin
findAndValidateName	Validate	Find & validate person name	Yes	name
findAndValidatePhone	Validate	Find & validate phone number	Yes	phone
findAndValidateUltimateParent	Recommend / Validate	Find & validate ultimate parent	Yes	ultimate-parent
findAndValidateWebsite	Recommend / Validate	Find & validate website	Yes	website

Result polling

MCP tool	Objective	Description	Live lookup	API reference
getEmailResult	Recommend / Validate	Get email validation result	No	email
getLegalIDResult	Recommend / Validate	Get legal ID validation result	No	legal-id
getLinkedinResult	Recommend / Validate	Get LinkedIn validation result	No	linkedin
getNameResult	Validate	Get name validation result	No	name
getPhoneResult	Validate	Get phone validation result	No	phone
getUltimateParentResult	Recommend / Validate	Get ultimate-parent result	No	ultimate-parent
getWebsiteResult	Recommend / Validate	Get website validation result	No	website

Live lookup flags tools that trigger an external lookup against third-party data sources (equivalent to MCP's openWorldHint=true). All tools are read-only and safe to retry — re-submitting a job mints a fresh job_id with no side effects.

Tool behavior

Most Delpha tools follow an asynchronous submit → poll pattern:

1. The agent calls a findAndValidate* tool with the input data. The server returns a job_id immediately.

2. The agent calls the paired get*Result tool with that job_id to fetch the result. While the job is still running, the result tool returns status_code=202 and the agent retries; on completion it returns status_code=200 with the full result envelope.

Example pair: findAndValidateEmail → getEmailResult. Typical end-to-end latency is a few seconds.

Every result envelope shares the same shape: { message, status, status_code, result, process_time }. The result object is entity-specific — see the linked API page for each tool for the exact fields and scoring semantics.

Feedback

We're actively adding tools and refining descriptions based on how agents actually use them. If you have feedback, hit a rough edge, or want to see a new capability, email us at [e-mail].

---

Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the ask query parameter:

GET https://help.delpha.io/delpha-api-and-mcp/mcp.md?ask=<question>

The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

We're actively adding tools and refining descriptions based on how agents actually use them. If you have feedback, hit a rough edge, or want to see a new capability, email us at [e-mail].