The enterprise signal layer
for every LLM workflow
The universe of Autobound enterprise signal data at your fingertips. Connect Claude Code, Cursor, Windsurf, or any MCP-compatible tool to 700+ B2B signals through a single MCP connection. The data layer for the AI-native GTM stack.
{
"mcpServers": {
"autobound-signals": {
"command": "npx",
"args": ["-y", "@autobound-ai/mcp-server"],
"env": {
"AUTOBOUND_API_KEY": "your_api_key"
}
}
}
}Quick install:
npx -y @autobound-ai/mcp-serverHow it works
Three steps to AI-native signal access
No SDKs to install. No endpoints to memorize. Just configure and start asking questions.
01
Get your API key
Create your free account and grab an API key. Takes under 60 seconds — no sales call needed.
02
Configure MCP
Add the Autobound MCP server to your claude_desktop_config.json, Cursor settings, or any MCP-compatible client.
03
Start querying
Ask your AI assistant to search signals, enrich companies, or filter by signal type. It handles the API calls for you.
Available Tools
14 tools. Full signal coverage.
Each tool maps to a core API capability. Your AI assistant picks the right one based on your question.
signal_search
Structured search across 700+ signal types. Filter by signal type, domain, date range, and more to find exactly the signals you need.
company_enrich
Pass a domain. Get back every signal: financial filings, hiring trends, news, tech stack changes, social activity. Bulk enrichment available for up to 100 domains.
contact_enrich
Returns contact-level signals (job changes, LinkedIn posts, milestones) and their employer's company signals. Bulk enrichment available.
company_timeline
Chronological view of all recent signals for a company. See what happened in the last 7, 30, or 90 days.
contact_timeline
Full signal timeline for a contact by email. Batch endpoint available for multiple contacts in one call.
signal_types
List all available signal types with descriptions, categories, and delivery stats. Useful for discovering what data is available.
account_info
Check credits, usage, API request logs, and signal database statistics. Monitor your consumption in real time.
Works with
The B2B data layer for every AI platform
Anyone building with LLMs can now plug into Autobound's signal infrastructure. AI SDR platforms, internal copilots, research agents, or your own IDE workflow.
Use Cases
Signal data where you already work
No context switching. No dashboard hopping. Just ask.
Research a prospect before outreach
Ask your AI assistant to pull the latest signals on a company before a call. Get funding rounds, leadership changes, and product launches without switching tabs.
Enrich a lead list in your IDE
Working with a CSV of prospects? Enrich each row with company and contact signals directly from Cursor or Claude Code. No API client needed.
Monitor target account signals
Ask Claude to check what's new with your top accounts. Get a daily briefing on hiring changes, earnings, news, and social activity, all from your chat interface.
Learn more
Guides and blog posts about building AI-native workflows with signals.
Signal API Reference
Full REST endpoint documentation — search, enrich, and monitor signals.
Developer Hub & Quickstart
Step-by-step guide to making your first API call, with code samples.
OpenAI Function Calling
Add signal lookup to GPT-4 agents via OpenAI function calling.
Claude Code Integration
Use signals directly from Claude Code with zero setup beyond an API key.
MCP Server Launch
Announcing Autobound's MCP server — how it works and why it matters for AI workflows.
The Activation Moat
How GTM teams win when AI thinks in context — the philosophy behind signal-powered agents.
FAQ
Frequently asked questions
What is MCP and how does it work with Autobound?
MCP (Model Context Protocol) is an open standard that lets AI tools like Claude, Cursor, and Windsurf connect to external data sources. Autobound's MCP server exposes 14 tools that give your AI assistant direct access to 700+ B2B signal types across 250M+ contacts and 50M+ companies.
Which AI tools support the Autobound MCP server?
Any MCP-compatible client works, including Claude Desktop, Claude Code, Cursor, Windsurf, Zed, and custom MCP hosts. The server uses the standard stdio transport, so any tool that supports MCP can connect.
Do I need to write code to use the MCP server?
No. You just add a JSON config block to your AI tool's settings file and provide your API key. After that, you can ask your AI assistant natural language questions like 'What signals fired for Salesforce this month?' and it handles the API calls automatically.
What's the difference between the MCP server and the REST API?
The REST API is for programmatic access — you write code that calls endpoints directly. The MCP server wraps that same API into AI-friendly tools so LLMs can call it autonomously. Same data, different interface. Use REST for production pipelines, MCP for interactive AI workflows.
Is there a cost to use the MCP server?
The MCP server itself is free and open-source (npm package). You pay for API usage based on your Autobound plan — same pricing whether you call via MCP, REST, or any other method. Start with 100 free credits.
Start building with signal data
Sign up, configure the MCP server, and start querying 700+ signals from your AI assistant.