Your Signal Data, One API Call Away

We've spent the last four years building the most comprehensive real-time signal engine in B2B. Some of the biggest names in CRM, sales intelligence, and revenue tech already rely on our data to power their platforms — we process SEC filings, LinkedIn posts, hiring trends, Reddit threads, Glassdoor reviews, and 20+ other signal types across hundreds of millions of contacts and companies.

Until now, getting that data meant GCS buckets, flat files, and a data engineering team to stitch it all together. It worked. But it wasn't exactly elegant.

Today we're launching the Autobound Signal API (beta) — and it changes everything about how teams consume signal intelligence.

The Old Way Was Fine. This Is Better.

Here's the honest truth: our existing customers are happy with bulk delivery. They get JSONL and Parquet drops on a schedule, match against their CRM, and move on. It works at scale.

But I kept hearing the same thing from prospects and smaller teams: "I don't want to spin up a data pipeline. I just want to ask a question and get an answer."

Fair enough.

The Signal API is that answer. One endpoint. Structured, enriched, real-time signal data — searchable in plain English or with hard filters. No ETL. No data engineering overhead. Just intelligence, on demand.

What You Can Do

Semantic Search

This is the one that gets people excited. You can search our entire signal corpus in natural language:

curl -s "https://signals.autobound.ai/v1/signals/search" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "companies expanding sales teams",
    "signal_types": ["hiring-trends", "hiring-velocity"],
    "employee_count_min": 100,
    "detected_after": "2026-02-01",
    "limit": 20
  }'

"Companies investing in AI infrastructure." "SaaS startups that just raised Series B." "Enterprises with declining Glassdoor ratings and rising engineering headcount." You combine natural language with structured filters — signal type, company size, date range — and get back clean, entity-resolved JSON.

This isn't keyword matching. Queries are embedded as vectors and matched against our signal summaries. It finds what you mean, not just what you typed.

Company & Contact Enrichment

Pass a domain, get back everything we know:

curl -s "https://signals.autobound.ai/v1/companies/enrich" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain": "salesforce.com"}'

Financial filings, hiring velocity, news events, tech stack changes, Reddit sentiment, leadership moves — every signal type, one response. Same thing works at the contact level with email or LinkedIn URL.

Timelines

Want a chronological view of what's been happening at a company?

curl -s "https://signals.autobound.ai/v1/companies/salesforce.com/timeline" \
  -H "x-api-key: YOUR_API_KEY"

Last 7, 30, or 90 days of signals, ordered by detection date. Perfect for account research before a call, or for feeding context into an AI workflow.

Real-Time Webhooks

For teams that want signals pushed, not pulled:

curl -s "https://signals.autobound.ai/v1/webhooks" \
  -X POST \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://your-app.com/signals",
    "signal_types": ["news", "funding-round", "leadership-change"],
    "secret": "whsec_your_signing_secret"
  }'

Subscribe to the signal types you care about, and we'll POST them to your endpoint as they're detected. Signed payloads, retry logic, the works. Wire it to your CRM, Slack, data warehouse — whatever.

The MCP Thing (This Is the Big One)

Here's where it gets interesting.

If you've been paying attention to what Anthropic is doing with the Model Context Protocol, you know that MCP is becoming the standard way AI agents interact with external tools and data. Claude, Cursor, Windsurf, Claude Code — they all speak MCP natively.

We built an MCP server for the Signal API. Which means you can do this:

{
  "mcpServers": {
    "autobound-signals": {
      "command": "npx",
      "args": ["-y", "@autobound/mcp-server"],
      "env": {
        "AUTOBOUND_API_KEY": "your_api_key"
      }
    }
  }
}

Drop that into your Claude Desktop config, and suddenly your AI assistant has direct access to real-time B2B signal intelligence. No browser tabs. No copy-pasting. No context-switching.

Ask Claude: "What hiring signals has Salesforce shown in the last 90 days?" It calls the API, gets structured results, and reasons about them in context. Ask it to "Find companies in fintech that just had leadership changes and are hiring for sales roles" — it composes the right filters, searches our corpus, and gives you a synthesized answer.

This is the thing I'm most excited about. We're not just building an API — we're building the intelligence layer that AI agents use to understand what's happening in the business world in real time. And as far as I know, we're the first signal data provider to ship a production MCP server.

The future of GTM isn't a dashboard. It's an agent that knows what's happening at your target accounts before you do, and takes action. The Signal API + MCP is step one.

26 Signal Types, 400+ Subtypes

Quick rundown of what's in the engine:

Financial Intelligence — 10-K, 10-Q, 8-K, earnings transcripts, financial fundamentals. We process every SEC filing with LLMs trained on document structure, extracting 70+ structured subtypes like aiInvestment, costReduction, internationalExpansion.

Social Intelligence — LinkedIn posts, Glassdoor reviews, GitHub activity, Reddit mentions, Twitter/X, YouTube. Each with NLP-extracted pain points, initiatives, technologies, and sentiment scores.

Workforce Intelligence — Hiring velocity, hiring trends, job changes, employee growth by department, career milestones. We track hiring across 21M+ domains.

Market Intelligence — News events, SEO & traffic trends, website changes, tech stack detection, patents, product reviews. 3,600+ distinct website change types monitored across ~2M company sites.

Every signal follows a universal schema with entity resolution (99%+ domain coverage, 90%+ LinkedIn URL coverage), confidence scores, and LLM-generated summaries.

Pricing

Credit-based, usage-driven. Every API call consumes credits based on query complexity — a simple enrich call costs less than a semantic search across millions of signals. We designed it so you can start small and scale without renegotiating contracts or talking to sales every time you want to try something new.

Beta pricing is available now. We're being generous with credits during beta because we want you building on this.

Get Started

1. Sign up at signalapi.autobound.ai/dashboard — get your API key instantly, no approval queue
2. Read the docs at autobound-api.readme.io — full schema reference, quickstart guide, code examples
3. Make your first call — zero to data in under 5 minutes
4. Try the MCP server — npx -y @autobound/mcp-server and connect to Claude or Cursor

We've been the signal engine behind some of the biggest GTM platforms in the world. Now that engine has an API. I genuinely think this changes how teams find and act on buying signals — and with MCP, how AI agents do it too.

Go build something. I can't wait to see what you ship.

— Daniel Wiener, CEO & Co-Founder