MCP Server

Connect Encrata to any AI agent using the Model Context Protocol.

Overview

Encrata provides an MCP (Model Context Protocol) server that lets any compatible AI agent — Claude, ChatGPT, Cursor, Windsurf, VS Code Copilot — use Encrata as a tool natively. Ask your AI assistant to look up a person by email, and it calls Encrata directly.

There are two ways to connect:

Method	Transport	Best for
npm package (recommended)	stdio	Claude Desktop, ChatGPT, Cursor, VS Code, Windsurf
Remote endpoint	Streamable HTTP	Custom integrations, any MCP client

Quick start

1. Get an API key

Sign up at encrata.com and create an API key in Settings → API Keys.

2. Install

npm install -g encrata-mcp

Or use npx — no install needed (shown in configs below).

Authentication

Pass your Encrata API key as an environment variable (npm package) or Bearer token (remote endpoint):

# npm package
ENCRATA_API_KEY=your-api-key

# Remote endpoint
Authorization: Bearer YOUR_API_KEY

Available tools

Tool	Description	Credits
lookup_email	Full person intelligence from an email — name, company, role, socials, breaches, validity	1 per fresh lookup
validate_email	Quick deliverability check — valid, invalid, or disposable	Free
check_breaches	Data breach exposure — affected services and exposed data types	Free
ip_lookup	IP intelligence — geolocation, VPN/proxy/Tor detection, ASN, company	Free
company_search	Company intelligence — profile data, knowledge graph, organic search results	1
domain_search	Domain lookup — WHOIS data, registrar, nameservers, knowledge graph	1
darkweb_search	Dark web intelligence — search by email, domain, IP, or free text	1
google_search	Google search — web, news, images with country/language filters	1
list_monitors	List all email monitors	Free
create_monitor	Create a new monitor to track email changes over time	Free
get_monitor	Get details of a specific monitor	Free
trigger_monitor_run	Trigger an immediate enrichment run	1 per email
list_runs	List monitoring runs for a monitor or all monitors	Free
get_run_results	Get enrichment results and detected changes for a run	Free
list_contact_lists	List all contact lists	Free
create_contact_list	Create a reusable email list	Free
get_contact_list	Get details of a specific contact list	Free
delete_contact_list	Delete a contact list permanently	Free
list_contact_list_emails	List all emails in a contact list	Free
add_emails_to_list	Add emails to a contact list	Free
remove_emails_from_list	Remove emails from a contact list	Free
list_bulk_jobs	List async bulk enrichment jobs	Free
get_bulk_job	Get status and download URL of a bulk job	Free
cancel_bulk_job	Cancel a pending/in-progress bulk job	Free

lookup_email

Parameter	Type	Required	Description
email	string	Yes	Email address to look up
fields	string	No	Comma-separated fields to return (see field reference)

validate_email

Parameter	Type	Required	Description
email	string	Yes	Email address to validate

check_breaches

Parameter	Type	Required	Description
email	string	Yes	Email address to check

ip_lookup

Parameter	Type	Required	Description
ip	string	Yes	IP address to look up (IPv4 or IPv6)

company_search

Parameter	Type	Required	Description
query	string	Yes	Company name or search query

domain_search

Parameter	Type	Required	Description
query	string	Yes	Domain name to look up (e.g. example.com)

darkweb_search

Parameter	Type	Required	Description
query	string	Yes	Search query (email, domain, IP, or text)
type	string	No	Type of search: email (default), domain, ip, or search
offset	number	No	Pagination offset for results

google_search

Parameter	Type	Required	Description
query	string	Yes	Search query
type	string	No	Type of search: search (default), news, or images
country	string	No	Country code for localized results (e.g. us, uk, in)
lang	string	No	Language code (e.g. en, fr)
num	number	No	Number of results to return

Connect to your AI tool

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

npm package (recommended)

{
  "mcpServers": {
    "encrata": {
      "command": "npx",
      "args": ["-y", "encrata-mcp"],
      "env": {
        "ENCRATA_API_KEY": "your-api-key"
      }
    }
  }
}

Then ask Claude: “Look up [e-mail] on Encrata”

Claude Code

claude mcp add encrata -- npx -y encrata-mcp

# Set your API key
export ENCRATA_API_KEY="your-api-key"

ChatGPT

ChatGPT supports MCP servers via the desktop app. Go to Settings → MCP Servers → Add:

• Name: Encrata
• Command: npx -y encrata-mcp
• Environment: ENCRATA_API_KEY=your-api-key

Cursor

Add to .cursor/mcp.json in your project root:

npm package (recommended)

{
  "mcpServers": {
    "encrata": {
      "command": "npx",
      "args": ["-y", "encrata-mcp"],
      "env": {
        "ENCRATA_API_KEY": "your-api-key"
      }
    }
  }
}

VS Code Copilot

Add to .vscode/mcp.json in your project root:

npm package (recommended)

{
  "servers": {
    "encrata": {
      "command": "npx",
      "args": ["-y", "encrata-mcp"],
      "env": {
        "ENCRATA_API_KEY": "your-api-key"
      }
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

npm package (recommended)

{
  "mcpServers": {
    "encrata": {
      "command": "npx",
      "args": ["-y", "encrata-mcp"],
      "env": {
        "ENCRATA_API_KEY": "your-api-key"
      }
    }
  }
}

Any MCP Client

npm package (stdio):

Command: npx -y encrata-mcp
Transport: stdio
Env: ENCRATA_API_KEY=your-api-key

Remote endpoint (HTTP):

URL: https://encrata.com/mcp
Transport: Streamable HTTP
Auth: Bearer token (your API key)

Example prompts

Once connected, try asking your AI:

• “Look up [e-mail]”
• “Is [e-mail] a valid email?”
• “Check if [e-mail] has been in any data breaches”
• “Find me everything about the person behind [e-mail]”
• “Validate these emails and tell me which are deliverable: [e-mail], [e-mail]”
• “Search for information about OpenAI as a company”
• “Look up the domain tesla.com — who owns it and when does it expire?”
• “Search the dark web for any mentions of [e-mail]”
• “Google search for ‘AI agent frameworks 2025’ and summarize the top results”
• “Find dark web mentions of the domain example.com”
• “Search Google News for ‘Series A funding’ in the US”

How it works

┌─────────────────────────────────────────────┐
│           AI Agent (Claude, ChatGPT, etc.)  │
└─────────────┬───────────────────────────────┘
              │ MCP protocol
              ▼
┌─────────────────────────────────────────────┐
│   npm package (stdio)    or    Remote HTTP  │
│   encrata-mcp                  /mcp         │
└─────────────┬───────────────────────────────┘
              │ REST API / JSON-RPC
              ▼
┌─────────────────────────────────────────────┐
│             Encrata API                     │
│   10+ data sources in parallel              │
└─────────────────────────────────────────────┘

npm package: Runs locally as a stdio process. Calls Encrata’s REST API directly. No sessions, no state — each tool call is a simple HTTP request.

Remote endpoint: Streamable HTTP transport (JSON-RPC 2.0 over POST) at https://encrata.com/mcp. Useful for custom integrations or clients that only support HTTP transport.

Source code

The MCP server is open source:

https://github.com/Encratahq/encrata-mcp

Advanced: raw JSON-RPC

If you’re building a custom integration against the remote endpoint:

Initialize

curl -X POST https://encrata.com/mcp \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "initialize",
    "params": {
      "protocolVersion": "2025-03-26",
      "capabilities": {},
      "clientInfo": { "name": "my-app", "version": "1.0" }
    }
  }'