enrich_personLABSCO SUMMARY
Prospect research answered where the email is being written.
Take the hosted path: a URL, one OAuth approval, and no key stored in the agent's config. The flow it supports is the short one — find the person, then get their details — done without leaving the draft.
What it is
An MCP server for DataLayer, a B2B data enrichment API covering companies and the people who work at them. A hosted remote server is the recommended path; the npm package is the stdio fallback for environments that cannot reach it.
What you get
- A person looked up by who they are and where they work
- A person enriched with verified contact details
- A commercial database of 60M+ companies and 300M+ verified contacts, also reachable through people and company search, a company's decision-makers, its tech stack, department headcount, open roles and buying-intent signals
- A hosted endpoint at https://api.datalayer.sh/mcp over Streamable HTTP, with OAuth and Dynamic Client Registration
- A stdio npm package, still maintained, for environments with no route to a remote server
Requirements
For the hosted server, paste the URL into your client and approve the OAuth prompt — no API key sits in the agent. Headless callers send an Authorization: Bearer header on POST /mcp instead.
Setup effort
Paste a URL, then sign in — add the endpoint to your client and sign in to the vendor
USE CASES
01
Finding the right person at a target company
Look them up by role and company instead of scrolling a professional network.
02
Getting a verified work email before you write
Enrich the person you found and read their contact details back.
03
Checking a lead is real before it goes into the CRM
Run the name through lookup and see whether it resolves to anyone.
04
Filling in a partial contact record
Hand over what you already have and get the rest back enriched.
11 tools
Listed in project docs. Your client may show a different list.
lookup_personsearch_peoplecompany_employeesenrich_companylookup_companysearch_companiescompany_technographicscompany_headcountcompany_jobsfind_intent_signalsBEFORE YOU INSTALL
What to know first
Tools you get
11 tools named in the project's documentation.
Requires
A client that can reach a remote MCP server over Streamable HTTP; A DataLayer account (free sign-up): the hosted server signs you in over OAuth; the npx package reads DATALAYER_API_KEY instead
INSTALLATION
Connect your client
Add the endpoint to your client and sign in to the vendor.
Server setup
ENDPOINT URL
https://api.datalayer.sh/mcpMCP Clients
Claude.ai
Settings → Connectors → Add custom connector → paste URL → click Approve
Claude Code
CLAUDE CODE · TERMINAL
claude mcp add datalayer https://api.datalayer.sh/mcp --transport httpCursor
CURSOR · JSON
{
"mcpServers": {
"datalayer": {
"url": "https://api.datalayer.sh/mcp"
}
}
}Claude Desktop, Windsurf
The stdio package, for clients that cannot reach the hosted server. Replace sk_live_your_key_here with the key from Dashboard → API Keys at datalayer.sh.
JSON
{
"mcpServers": {
"datalayer": {
"command": "npx",
"args": [
"-y",
"@datalayer-sh/mcp"
],
"env": {
"DATALAYER_API_KEY": "sk_live_your_key_here"
}
}
}
}Headless — bearer token
Authorization: Bearer sk_live_... on POST /mcp
