DelphaMCP

Delpha Data Quality MCP integrates Delpha's AI-powered data quality tools into your MCP environment. It exposes a suite of data quality tools (email, address, website, LinkedIn, phone, name, and legal ID) that can be invoked from within your AI assistant workflows to validate, normalize, enrich, and deduplicate customer data. The MCP wires the Delpha modules into your existing pipeline so you can request actions such as findAndValidateEmail, findAndValidateAddress, findAndValidateWebsite, and similar tool names, receiving normalized results, confidence scores, and recommended improvements. To use it, configure the Delpha MCP server with OAuth2 credentials, restart your application, and then invoke the available tools through your MCP prompts or automation flows. The server is designed to help keep CRM and contact data clean, up-to-date, and consistent across sources while reducing manual data cleaning effort.

Prerequisites:\n- Python 3.8+ installed on your system.\n- Access to install Python packages from PyPI.\n- An MCP-compatible application container or runtime to load the Delpha MCP server.\n\nInstallation steps:\n1) Install the Delpha MCP package:\n pip install delpha-mcp\n\n2) Obtain OAuth2 credentials from Delpha:\n - Client ID and Client Secret (contact support.api@delpha.io if you don't have them)\n\n3) Configure your MCP to load the Delpha MCP server:\n Create or update your MCP configuration with the Delpha server entry:\n\n {\n "mcpServers": {\n "Delpha": {\n "command": "python",\n "args": ["-m", "delpha_mcp"],\n "env": {\n "DELPHA_CLIENT_ID": "your_client_id_here",\n "DELPHA_CLIENT_SECRET": "your_client_secret_here"\n }\n }\n }\n }\n\n4) Restart your MCP-enabled application to apply the changes. You should now be able to call Delpha tools from your assistant or workflows.\n

Notes and tips:\n- Ensure your DELPHA_CLIENT_ID and DELPHA_CLIENT_SECRET are kept secure and not committed to source control.\n- If you encounter authentication issues, verify that the OAuth2 credentials have the appropriate scopes for data access.\n- The Tools include Email, Address, Website, LinkedIn, Phone, Name, and Legal ID validators and normalizers. Each tool provides completeness, validity, accuracy, and consistency checks, along with optional AI-driven recommendations.\n- For best results, supply context (e.g., country, company, or domain) where relevant to improve normalization and disambiguation.\n- If you plan to run multiple MCP servers, ensure unique server names to avoid conflicts.