---
title: "Funding signals in your AI agent with MCP | Datahyena"
url: https://datahyena.com/blog/funding-signals-in-your-ai-agent-with-mcp/
description: "Connect funding and company signals to your AI agent with one config line, then ask in plain language. What MCP is, the tools, and a worked example."
---

[← Back to blog](https://datahyena.com/blog) mcp api

# Funding signals in your AI agent with MCP

 Connect funding and company signals to your AI agent with one config line, then ask in plain language. What MCP is, the tools, and a worked example.

 Akash Rajpurohit · July 28, 2026 · 8 min read
 ![Funding signals in your AI agent with MCP](https://datahyena.com/static/images/scenaries/scenary-028.png)

 You can put funding and company signals inside your AI agent with one line of config. After that you stop copy-pasting from a dashboard and start asking your agent for the rounds, acquisitions, and exec moves you care about in plain language.

This guide covers what MCP is, how connecting works, the tools your agent gets, and a worked example of a question you can ask and what comes back.

## TLDR

- MCP lets an AI agent call external tools. Datahyena exposes its signals as MCP tools, so your agent can query them directly.

- Connecting is a single config line in your MCP client, plus a one-time browser authorization.

- Your agent gets six read tools: funding events, acquisitions, executive moves, companies, investors, and a single-call company timeline.

- You ask in plain language. The agent picks the tool, applies the filters, and hands back clean, typed records.

- It is the same data as the REST API, billed the same way: one credit per record.

## What is MCP?

MCP, the Model Context Protocol, is an open standard that lets an AI agent call external tools instead of guessing from training data. Datahyena exposes its signals as MCP tools, so your agent can query live funding and company data on its own.

That is the whole idea. Your agent already knows how to call tools. MCP is the standard way to hand it a new one. You do not build a plugin or wrap an API. You point a compatible client at the server and the tools show up.

Any MCP client works: Claude, Cursor, Codex, Gemini, and anything else that speaks the protocol. The agent sees typed inputs and structured outputs, not a screen to scrape.

## How do you connect an AI agent to funding data?

You connect by adding the remote MCP server URL in your client, then approving a one-time browser authorization. That is the entire setup.

For most clients it is a single line in a config file:

```
{
 "mcpServers": {
 "datahyena": { "url": "https://api.datahyena.com/mcp" }
 }
}
```

The first time your agent reaches the server, your browser opens. You sign in, pick a workspace, and approve read access. Auth is OAuth 2.1 with PKCE, the token is scoped to that workspace, and you can revoke it from the dashboard any time.

After that, the tools are available in every session. You never paste an API key into a chat window.

## What tools does the agent get?

Your agent gets six read tools. Each maps to a public API endpoint, with the same filters and the same typed response.

| Tool | What your agent can ask for |
| --- | --- |
| funding_events | New funding rounds with company, investors, amount, and round. Filter by round, amount, country, or industry. |
| acquisitions | Acquisitions and mergers with acquirer, target, deal value, and payment type. |
| exec_moves | Executive appointments, promotions, departures, and transitions with person, company, and seniority. |
| companies | Company records. Search by name, domain, country, industry, employee bucket, or founded year. |
| investors | Investor records. Search by name, type, country, or domain. |
| company_timeline | One company’s full timeline: funding, acquisitions, and executive moves, plus firmographics and investors. Resolve by id, domain, name, or LinkedIn URL. |

Notice these are read-only and they line up one-to-one with the [funding API](https://datahyena.com/blog/funding-data-api-guide?utm_source=marketing&utm_medium=blog&utm_campaign=funding-signals-in-your-ai-agent-with-mcp) and the rest of the signal endpoints. The agent reads signals. It does not write anything to your workspace.

## A worked example

Here is what it looks like in practice. You ask your agent a normal question, and it calls the right tool with the right filters.

The prompt:

```
Show me Series A and B fintech rounds from the last week,
US and UK only, above $10M. List company, amount, and lead investor.
```

The agent picks `funding_events`, applies the filters, and gets back resolved records. The shape it works from looks like this:

```
{
 "data": [
 {
 "id": "fund_01J9X8…",
 "company": {
 "id": "co_01J7A2…",
 "name": "Acme Payments",
 "domain": "acme.com",
 "industry": "fintech",
 "hq": "London, GB"
 },
 "round": "series_b",
 "amount": 24000000,
 "currency": "USD",
 "announced_date": "2026-08-10",
 "investors": [
 { "id": "inv_01H3…", "name": "Northwind Capital", "role": "lead" }
 ]
 }
 ]
}
```

Then your agent reads that back to you in plain language: Acme Payments raised a $24M Series B on August 10, led by Northwind Capital, plus the others that match.

Because every field is already typed, you can keep going without leaving the chat. Ask “which of these has no known head of sales yet” and the agent calls `exec_moves` and `companies` next. The amounts are integers it can compare, the round labels are normalized values it can group, and the company IDs are stable, so it can chain one answer into the next.

## What can you ask once it’s connected?

Once the tools are live, you ask the way you would ask a teammate who watches the market all day. The agent maps your question to a tool and the right filters, then reads the records back.

Here are questions that map cleanly to the six tools:

- “Which fintech companies raised a Series A in the last 30 days?” The agent calls `funding_events`, filters by round, industry, and date, and lists the companies with amounts and leads.

- “Who are the most active investors in climate this quarter?” It pulls recent `funding_events` in that sector and counts which investors show up most often.

- “Did anyone acquire a company in our space last week?” It calls `acquisitions`, filters to your industry and the date range, and returns acquirer, target, and deal value.

- “What exec moves happened at our target accounts?” It calls `exec_moves` for the companies you name and lists the new roles, the people, and the seniority.

- “Pull every company we track in devtools founded after 2020 with a known domain.” It calls `companies` and returns the records ready to join to your own data.

You do not learn a query language for any of this. You ask in plain words, and the agent does the filtering. If the first answer is close but not exact, you refine in the next message and it re-runs the tool.

## How does this fit an agentic workflow?

It fits because the agent can do something with the answer, not just show it to you. The signal is the start of the work, not the end.

A common loop looks like this. The agent queries `funding_events` for rounds that match your market. For each company it pulls the matching `exec_moves` and `companies` record so it knows who just joined and what the firmographics are. Then it drafts the outreach, or updates a record in the tool you already work in, or flags the account for a human to review.

Because the records are typed and the IDs are stable, the agent can chain these steps without you copy-pasting between screens. It compares amounts, groups by round, and carries the same company ID from the funding answer into the exec-move lookup. One question becomes a short pipeline of actions.

You stay in control of what happens next. The tools only read signals, so nothing is written to your workspace by the server. The agent proposes the draft or the update, and you decide whether to send it.

## Is this the same data as the API?

Yes. The MCP tools return the same resolved records as the REST API, and they are billed the same way.

There is one source of signals behind everything. The cleaning, deduplication, and entity resolution happen once. The REST API, webhooks, and MCP are three ways to read the same record, not three different datasets. A round you see over MCP is the same round you would get from the API.

Billing is shared too. One credit per record returned, drawn from the same balance, on every plan including the free tier. There is no separate MCP subscription to manage and no second meter to watch.

So the choice between MCP and the API is about how you want to consume, not what you get. MCP is the conversational front door for asking questions and exploring. The [REST API](https://datahyena.com/api?utm_source=marketing&utm_medium=blog&utm_campaign=funding-signals-in-your-ai-agent-with-mcp) and webhooks are for production pipelines that pull or receive signals on a schedule.

## When is MCP the right way in?

MCP is the right way in when a person or an agent is asking questions, not when a system is moving records on a schedule.

Reach for MCP when you want to explore signals in natural language, when you are building an agent that needs live company data mid-conversation, or when you want your team querying signals from the tools they already work in. The friction is near zero: one config line and you are asking questions.

Reach for the REST API or webhooks when you need signals flowing into a database, a CRM, or a job that runs every hour. Those are for throughput and automation, where you want a fixed contract rather than a conversation.

Most teams use both. They wire the API into the pipeline and keep MCP open for the ad-hoc questions that come up day to day. Same records, same credits, two doors.

## Start asking your agent

The fastest way to understand this is to connect once and ask a real question. Add the [MCP server](https://datahyena.com/mcp?utm_source=marketing&utm_medium=blog&utm_campaign=funding-signals-in-your-ai-agent-with-mcp) to your client, authorize in the browser, and ask for a funding round you already know about to see the clean record come back. You get 50 free credits, no card required.

## Frequently asked questions

 What is the MCP server for funding signals? It is a Model Context Protocol server that exposes Datahyena's signals as tools an AI agent can call. Your agent can query funding rounds, acquisitions, executive moves, companies, and investors in plain language and get back clean, typed records.
 How do I connect my AI agent to funding data over MCP? Add the remote MCP server URL in your MCP client, then approve a one-time browser authorization. That is the whole setup. After that, you ask your agent for the signals you need and it calls the tools for you.
 Is MCP access included in my plan? Yes. MCP is available on every plan, including the free tier. It uses the same credits as the REST API, one credit per record returned, so there is no separate billing to manage.
 Is MCP the same data as the REST API? Yes. The MCP tools return the same resolved records as the REST API, with the same filters and the same response shape. MCP is the conversational way in. The API and webhooks are for production pipelines.

Keep reading

## More from the blog

 [data-quality Aug 2, 2026 · Akash Rajpurohit

## Build vs buy funding data: the real cost

 Building a funding data pipeline is cheap to start and expensive to keep alive. Here's the honest build vs buy trade-off, with a comparison table.

Read post
→](https://datahyena.com/blog/build-vs-buy-funding-data-pipeline) [api Jul 31, 2026 · Akash Rajpurohit

## Webhooks vs polling for funding events

 When to use webhooks and when to poll for funding events, with a clear trade-off table on latency, reliability, replay, and complexity.

Read post
→](https://datahyena.com/blog/webhooks-vs-polling-for-funding-events) [data-quality Jul 23, 2026 · Akash Rajpurohit

## Entity resolution for company data

 One company shows up as Acme, Acme AI, Acme Inc, and acme.ai. Here's why they must resolve to a single record, and what breaks when they don't.

Read post
→](https://datahyena.com/blog/entity-resolution-for-company-data)

## Start pulling signals in minutes.

Create a key, claim your 50 free credits, and make your first request today. No sales call,
 no credit card.

 [Get your API key

→](https://app.datahyena.com/register) [Read the docs](https://datahyena.com/docs)

50 free credits · no credit card