Give your AI agent access to 60M+ companies and 300M+ verified contacts. Enrich leads, find work emails, discover tech stacks, and identify buying intent — directly from Claude, Cursor, Windsurf, or any MCP-compatible AI agent.
🚀 New: Hosted MCP server. The recommended way to connect to DataLayer is now the hosted remote MCP server at
https://api.datalayer.sh/mcp. It uses Streamable HTTP (MCP spec 2025-11-25) and OAuth 2.1 + Dynamic Client Registration — paste the URL into Claude.ai, click Approve, done. No install, no API key in the agent. See "Hosted MCP Server: One URL, OAuth, No Install" for the full setup. This npm package (stdio transport) is still maintained as a fallback for environments that can't reach a remote server.
URL: https://api.datalayer.sh/mcp
| Client | Setup |
|---|---|
| Claude.ai | Settings → Connectors → Add custom connector → paste URL → click Approve |
| Claude Code | claude mcp add datalayer https://api.datalayer.sh/mcp --transport http |
| Cursor | ~/.cursor/mcp.json → { "mcpServers": { "datalayer": { "url": "https://api.datalayer.sh/mcp" } } } |
| Headless | Authorization: Bearer sk_live_... on POST /mcp |
OAuth metadata (RFC 9728): https://api.datalayer.sh/.well-known/oauth-protected-resource
The rest of this README covers the legacy stdio package for environments where the hosted server isn't reachable.
DataLayer is a B2B data enrichment API. This MCP server lets AI agents call it directly — no code, no HTTP requests, just natural language.
Ask your agent:
"Find the VP of Engineering at Stripe and get their verified email"
"Show me all SaaS companies in the US using Salesforce with 50+ employees that are actively hiring"
"What's the tech stack at HubSpot? Do they use AWS or GCP?"
"Find companies spending $10K+/mo on Google Ads with employee growth above 15%"
The agent calls the right DataLayer tool, gets structured data back, and reasons about it.
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"datalayer": {
"command": "npx",
"args": ["-y", "@datalayer-sh/mcp"],
"env": {
"DATALAYER_API_KEY": "sk_live_your_key_here"
}
}
}
}Add to .cursor/mcp.json in your project root:
{
"mcpServers": {
"datalayer": {
"command": "npx",
"args": ["-y", "@datalayer-sh/mcp"],
"env": {
"DATALAYER_API_KEY": "sk_live_your_key_here"
}
}
}
}Same config pattern. Set the DATALAYER_API_KEY environment variable and run npx -y @datalayer-sh/mcp via stdio.
- Sign up at datalayer.sh (free, no credit card)
- Go to Dashboard → API Keys → Create key
- Copy the
sk_live_...key into your config
10 free credits included — enough to test every tool.
| Tool | What it does | Credits |
|---|---|---|
enrich_person |
Get name, verified email, phone, job title, seniority, LinkedIn URL for any contact | 1 |
lookup_person |
Find a person by email, phone number, LinkedIn URL, or name + company | 1 |
search_people |
Search 300M+ contacts by title, seniority, function, company, location | 1/result |
company_employees |
List decision-makers at any company, filtered by role and seniority | 1 |
Example: "Find all VPs of Engineering at Shopify with verified emails"
| Tool | What it does | Credits |
|---|---|---|
enrich_company |
Full company profile — industry, headcount, revenue, funding, tech stack, traffic, ad spend | 1 |
lookup_company |
Find a company by domain, LinkedIn URL, or name | 1 |
search_companies |
Search 60M+ companies by industry, size, tech stack, funding, traffic, growth | 1/result |
Example: "Show me Series A fintech companies in the US with 50-200 employees"
| Tool | What it does | Credits |
|---|---|---|
company_technographics |
Full tech stack across 16 categories — CRM, cloud, marketing automation, analytics, etc. | 1 |
company_headcount |
Department breakdown (engineering, sales, marketing, etc. — 15 departments) | 1 |
company_jobs |
Open roles by department — a hiring intent signal | 1 |
Example: "What CRM does Notion use? What about their marketing automation stack?"
| Tool | What it does | Credits |
|---|---|---|
find_intent_signals |
Companies ranked by buying intent — web traffic, Google ad spend, hiring velocity, employee growth, recent funding | 5/result |
Example: "Find companies in the SaaS industry that are spending heavily on Google Ads and growing fast"
- Build targeted lead lists: "Find CTOs at SaaS companies with 50-500 employees using AWS"
- Enrich CRM data: Pipe company + contact data into HubSpot or Salesforce
- Identify buying intent: Find companies actively hiring + spending on ads = ready to buy
- Competitor analysis: "What tech stack does [competitor] use? Who are their key hires?"
- Source candidates: "Find senior engineers at companies using React and TypeScript in San Francisco"
- Map org charts: Get headcount by department for any company
- Track hiring signals: Companies with many open roles = growing = recruiting opportunities
- Due diligence: Full company profile with funding history, growth rate, tech stack
- Deal sourcing: "Find recently funded AI companies with high employee growth"
- Market mapping: Search by industry, size, technology, and geography
- Automate prospecting: Build Claude Code pipelines that enrich leads automatically
- n8n / Make workflows: Trigger enrichment from any automation platform
- CRM enrichment bots: Agents that keep your CRM data fresh and complete
| Metric | Coverage |
|---|---|
| Companies | 60M+ worldwide |
| Contacts | 300M+ verified professionals |
| Emails | Verified with 90-day re-verification cycle |
| Phone numbers | Direct dials and mobile numbers |
| Tech stack | 16 categories (CRM, cloud, marketing, analytics, etc.) |
| Signals | Google Ad Spend, web traffic, employee growth, funding, hiring velocity |
| Geography | Global — US, EU, APAC, LATAM |
- 1 credit = 1 enrichment (company or person)
- Search results: 1 credit per result returned
- Intent signals: 5 credits per result (premium data)
- Failed lookups (404): Free — no credits consumed
- Credits never expire on top-up packs
10 free credits on signup. Plans from $49/mo.
| Variable | Required | Description |
|---|---|---|
DATALAYER_API_KEY |
Yes | Your API key from datalayer.sh |
DATALAYER_API_URL |
No | Override API base URL (default: https://api.datalayer.sh) |
| DataLayer | Apollo | Clay | ZoomInfo | |
|---|---|---|---|---|
| MCP server | Yes | Limited | No | No |
| REST API | Yes | Limited | No | Limited |
| Intent signals | Yes (unique) | Some | No | Yes |
| Pricing | From $49/mo | $49+/user/mo | $185+/mo | $6K+/yr |
| Credits expire | Never | Monthly | Monthly | Annual |
| Legal risk | Licensed data | OK | Varies | OK |
| Failed lookups | Free | Charged | Charged | Charged |
- Website
- Live Demo — try the data, no signup required
- API Documentation
- Dashboard & API Keys
Do I need to write code to use this? No. Install the MCP server, add your API key, and talk to your AI agent in natural language. The agent handles everything.
What AI tools does this work with? Any MCP-compatible client: Claude Desktop, Claude Code, Cursor, Windsurf, Cline, Continue, and more.
Is the data scraped from LinkedIn? No. DataLayer uses licensed, GDPR/CCPA-compliant data. No scraping, no legal risk. (ProxyCurl was shut down for scraping — we don't have that problem.)
How fresh is the data? Contacts are re-verified on a 90-day cycle. Company data is refreshed monthly.
What happens if a lookup returns no results? You're not charged. Failed lookups (404s) are free.
Issues and pull requests are welcome. Please open an issue first to discuss what you'd like to change.