Company Funding with MCP

Ask Claude or Cursor how much a company has raised and the agent calls get_company for you. One URL, OAuth sign-in, 1 record per company found.

This page is about setup and prompts. For the field list and REST call, see Company Funding. Other clients are on the MCP hub.

Connect the server

The server is https://api.leadocean.io/mcp. It signs in with OAuth 2.1, so there is no key to paste. Create a free account first.

Claude Code. The command format is from the Claude Code MCP docs, September 2026:

claude mcp add --transport http leadocean https://api.leadocean.io/mcp

Then run claude mcp login leadocean, or /mcp inside a session, and finish the browser sign-in.

Cursor. Add a remote server to mcp.json, in the url form shown in the Cursor MCP docs, September 2026:

{
  "mcpServers": {
    "leadocean": {
      "url": "https://api.leadocean.io/mcp"
    }
  }
}

Cursor starts the sign-in the first time the server is used. Other clients are listed in the LeadOcean MCP docs.

The tool

Funding has no tool of its own. It is one block inside the company record, returned by get_company (listed by the server as leadocean_get_company). It costs 1 record per company found.

Send one of the two. Name-only lookups do not work, so the agent asks for a domain. To find companies by funding stage instead, the agent calls search_companies with fundingType, minFunding, minLastFundingYear or investors. That is 1 record per company returned.

Prompts that work

Put a record limit in the prompt. "Stop after 25 records" is enough for the agent to hold a batch.

A question about reach goes to count_leads, which is free. A named company goes to get_company. A criteria list goes to search_companies.

The funding filters skip unknowns. A company with no figure on file is never returned by minFunding or minLastFundingYear, so a filtered list undercounts the real market. Say so in the prompt if the list feeds a decision.

What the agent gets back

The result is one leadocean.company.v1 object plus a meta block. The funding fields are from the company docs, September 2026. The docs sample response omits the funding block, so the values below are placeholders.

{
  "success": true,
  "data": {
    "company_data": {
      "company_name": "Example Inc",
      "company_domain": "example.com",
      "company_funding": {
        "total_raised_usd": 42000000,
        "last_round_type": "Series B",
        "last_round_date": "2025-06-12",
        "last_round_year": 2025,
        "lead_investors": ["Example Ventures"]
      }
    },
    "meta": { "sources": ["own"], "schema": "leadocean.company.v1" }
  },
  "meta": { "credits": 1, "source": "own" }
}

Funding is held for a small fraction of companies. Absent is the normal case, not an error. Values we do not hold come back null, never guessed. Tell the agent to write "not held" rather than fill the gap.

The record holds the last round only: its type, date and year, plus a total. There is no round-by-round history, and no alert when a company raises. This is a lookup, so rerun it when you need a fresh read.

Free checks first

count_leads, list_enum_values and get_account spend no records. Run them before any paid call.

The free count does not reliably apply the funding filters, so we do not quote a funding count here. Size the audience with filters it does apply, and the funding filter narrows the list at search time. For example, we ran count_leads on 1 October 2026 with jobLevel: ["C-Team"], hqCountry: ["US"], employeeRange: ["51-200"] and industry: ["Software Development"]. It returned total: 28793, totalIsExact: true, credits: 0. That counts people with a verified, catch_all_valid or catch_all email, not companies.

Troubleshooting

FAQ