// agentdatahub — data layer for AI agents

The data layer for AI agents.

Six datasets, one API key. Company data, contact data, contact verification, email finder, email verification, SERP API — all behind a single REST + MCP endpoint, with pooled credits across every call. Built for the developer wiring an autonomous agent, not for clicking around in a UI.

Request beta access Browse datasets

✓ REST + MCP-native — drop into Claude, Cursor, n8n, custom agents
✓ Pooled credits across every dataset — no separate vendor budgets
✓ Sub-100ms median latency on lookups
✓ Built on top of Clevenio's data infrastructure

api.agentdatahub.com

# Your agent calls one API. Six datasets behind it.
$ curl https://api.agentdatahub.com/v1/companies/example-company.com \
    -H "Authorization: Bearer $ADH_KEY"

{
  "domain": "example-company.com",
  "name": "Example Company",
  "country": "DE",
  "vertical": "Fintech",
  "employees": "1001-5000",
  "funding_stage": "Series E",
  "tech_stack": ["AWS", "React", "Stripe"],
  "signals": { "hiring": true, "tech_change_30d": false }
}

Datasets behind one key

MCP

Native + REST + SDKs

190+

Countries covered

// THE STACK

Six datasets behind one API key.

Stop integrating six vendors and reconciling six credit pools. agentdatahub bundles the data your agent actually needs — companies, contacts, verification, email, SEO — under one key.

🏢

Company data

Firmographics for any domain. Industry, size, funding, tech stack, signals.

👤

Contact data

Person-level records. Job title, seniority, department, location, social profiles.

✓

Contact verification

Real-time check that a contact still exists, in role, and reachable.

✉️

Email finder

Name + domain → work email. Or email → person. Forward and reverse, one endpoint.

🛡️

Email verification

Syntax, MX, SMTP, disposable, role-account checks. Deliverability scoring.

🔎

SERP API

SERP results, keyword volumes, backlinks, WHOIS — for agents that research the web.

See the MCP server

Or jump straight to pricing.

// WHY agentdatahub

One key, every dataset, agent-native.

Other data APIs were designed for a salesperson clicking through a UI, then bolted on an API later. agentdatahub starts from the opposite end: the agent calling the tool.

🔑

One API key, six datasets

Company data, contact data, contact verification, email finder, email verification, SERP API. Pooled credits across every endpoint — no separate budgets per vendor.

🤖

MCP-native

Native Model Context Protocol server. Drop the config into Claude Desktop, Cursor, Windsurf, or any MCP-compatible host and your agent can call every dataset as a tool.

⚡

Built for agent latency

Sub-100ms median on lookups, async streaming on bulk. Cached results are free for 24 hours. No 'enterprise demo' wall in front of throughput.

🧩

Stable, typed schemas

Every dataset has a documented schema. We version it. Your agent's tool definitions don't change under your feet.

🌍

Global coverage

190+ countries on company + contact data. Built on Clevenio's underlying infrastructure that already powers production data work in the Nordics, EU, and beyond.

💸

Priced like infra

One credit pool across every dataset, so you are not buying a separate plan per endpoint. Pricing is set at general availability — during the beta there is nothing to buy.

// HOW IT WORKS

From signup to first call in under 5 minutes.

Request beta access

Tell us what you're building. We're onboarding design partners now and rolling out broader access in waves.

Get your API key

One key unlocks every dataset. Hit the REST endpoint directly, or paste the MCP config into your agent host.

Wire it into your agent

Python, TypeScript, cURL, or MCP. Your agent now has company data, contacts, email tools, verification, and SERP — all on one credit pool.

// CODE

REST. MCP. SDKs. Whatever your agent speaks.

Same data, four shapes. Pick the one that fits the host you're shipping in.

curl https://api.agentdatahub.com/v1/companies/example-company.com \
  -H "Authorization: Bearer $ADH_KEY"

# returns 1 credit worth of company data:
{
  "domain": "example-company.com",
  "name": "Example Company",
  "country": "DE",
  "vertical": "Fintech",
  "employees": "1001-5000",
  "funding_stage": "Series E",
  "tech_stack": ["AWS", "React", "Stripe"],
  "signals": { "hiring": true, "tech_change_30d": false }
}

from agentdatahub import Client

adh = Client(api_key="...")

# 1. Find a contact's email by name + domain
email = adh.email_finder.find(
    first_name="Maximilian",
    last_name="Tayenthal",
    domain="example-company.com",
)
# {"email": "first.last@example-company.com", "confidence": 0.94}

# 2. Verify it's deliverable
status = adh.email_verification.verify(email["email"])
# {"deliverable": True, "score": 0.98, "disposable": False}

# 3. Pull the company record
company = adh.companies.get("example-company.com")

import { AgentDataHub } from "@agentdatahub/sdk";

const adh = new AgentDataHub({ apiKey: process.env.ADH_KEY });

// Same six datasets, fully typed
const company = await adh.companies.get("example-company.com");
const contact = await adh.contacts.find({
  domain: "example-company.com",
  title: "Head of Engineering",
});
const verified = await adh.contactVerification.check(contact.id);
const serp = await adh.serp.search({ q: "fintech licence germany", depth: 10 });

// ~/.claude_desktop/config.json or .cursor/mcp.json
{
  "mcpServers": {
    "agentdatahub": {
      "command": "npx",
      "args": ["-y", "@agentdatahub/mcp"],
      "env": { "ADH_KEY": "your-api-key" }
    }
  }
}

// Your agent now has tools:
//   adh_company_lookup     adh_contact_find
//   adh_contact_verify     adh_email_find
//   adh_email_verify       adh_serp_search

// PRICING

Pricing lands with general availability.

The API is in private beta. Plans and credit bundles are not final yet, so we are not publishing prices we might change. Request beta access, tell us what you are building, and we will talk through what it would cost for your volume.

Request beta access

// FAQ

Frequently asked questions

Is the API live yet? +

We're in early access — onboarding design partners now and rolling out broader access through Q2-Q3 2026. Request beta access and tell us what you're building; partners get hands-on help shaping the schema for their use case.

What's an MCP server? +

Model Context Protocol is the open standard for connecting AI agents to external tools. agentdatahub ships a native MCP server — drop one config block into Claude Desktop, Cursor, Windsurf, or any MCP-compatible host and every dataset becomes a tool your agent can call.

How is this different from Apollo, ZoomInfo, Clearbit, Hunter? +

Those products were designed for sales reps clicking around in a UI. They each cover one slice (Apollo = contacts + sequencing, Hunter = email finder, Clearbit = company enrichment) and they each charge separately. agentdatahub bundles the six datasets an agent actually needs behind one API key, with pooled credits, and is engineered as an API + MCP server first.

What sources does the data come from? +

Public web data: company websites, public business registers, funding databases, public professional profiles, job boards, and tech-detection signals. Built on top of Clevenio's underlying infrastructure that has indexed Nordic and EU markets at depth for years.

What's a credit? +

1 credit = 1 record returned. Cached re-fetches of the same record within 24 hours are free. Bulk endpoints and SERP queries have a per-record credit cost defined in the schema. Credits pool across every dataset — they're not split per-vendor.

Do I need to commit to a plan to test it? +

The API is in private beta, so there is no public price list yet and nothing to pay for during the beta. Request beta access and tell us your expected volume — we will walk through what pricing would look like at general availability.

GDPR, compliance, all that? +

We only ship publicly available data. Contact records include a provenance field on every response. EU customers can request EU data residency on Scale and above. We're operated by Clevenio Oy, a Finnish company.

// JOIN THE WAITLIST

Request beta access

Tell us about the agent you're building. Design partners get hands-on onboarding and an early API key within a few business days.

The data layer for AI agents.

Six datasets behind one API key.

Company data →

Contact data →

Contact verification →

Email finder →

Email verification →

SERP API →

One key, every dataset, agent-native.

One API key, six datasets

MCP-native

Built for agent latency

Stable, typed schemas

Global coverage

Priced like infra

From signup to first call in under 5 minutes.

Request beta access

Get your API key

Wire it into your agent

REST. MCP. SDKs. Whatever your agent speaks.

Pricing lands with general availability.

Frequently asked questions

Request beta access

Company data

Contact data

Contact verification

Email finder

Email verification

SERP API