Version 1.1.0 (updated October 2026)

Turn a messy company list into clean, structured lead profiles in minutes, without leaving your AI agent (Claude Desktop, Cursor, or any MCP client).

Lead Enrichment & Cleanup is an MCP (Model Context Protocol) server with one tool, enrich_leads. Paste a list of companies (JSON or CSV) and get back clean, de-duplicated lead profiles.

WHAT IT DOES

• Cleans company names: trims extra spaces, removes legal suffixes (Inc, Ltd, LLC, GmbH...), fixes ALL-CAPS and all-lowercase names

• Normalizes domains (lowercase, no https/www/path) and removes duplicates

• Flags invalid email syntax

• Rejects rows that have no company name and no valid domain

• AI classification with Google Gemini: inferred industry, company size band, key contact role, and a confidence score

• If one batch fails (rate limit, network), the other batches still return

EXAMPLE

Input: " Globex Corp ", "GLOBEX.COM"

Output: Globex, globex.com, plus industry, size band, contact role and confidence

IMPORTANT NOTE

The AI fields are estimates based only on the text you provide. This is not a verified firmographic database.

REQUIREMENTS

• Node.js 18 or newer

• Your own Google Gemini API key (free tier available at aistudio.google.com/apikey). Your lead data is sent to Google using your key.

• Any MCP client that supports stdio servers

WHAT'S INSIDE THE ZIP

• Complete MCP server code (src/index.js, src/lib.js)

• Editable classification prompt (prompts/enrich.txt)

• Sample input and output datasets

• manifest.json with a ready config snippet

• Offline self-test (npm test, no key needed)

• README with 3-step setup and troubleshooting

• LICENSE file

SETUP IN 3 STEPS

Step 1: Unzip and run: npm install

Step 2: Run: npm test (offline, no key needed)

Step 3: Add the server to your MCP client with your Gemini key