If you are building an AI agent or assistant that needs to research companies and leads mid-conversation, wiring up separate data sources is the slow part. A Model Context Protocol (MCP) server lets the agent call sales-intelligence tools directly, with no glue code.
Who uses a sales-intelligence MCP server
This tool is aimed at anyone building AI agents, assistants, or automations that need to research companies and people mid-task. Sales-engineering and RevOps teams wire it into agents that qualify and enrich leads. Founders and solo builders use it to give a personal assistant real lookup ability instead of guesswork. Automation teams add it to workflows that need company context at a decision point. In each case the point is the same: the agent gets structured intelligence without a human stopping to run a separate tool.
What it does
The Sales Intelligence MCP server exposes company- and lead-lookup tools to any MCP-capable client (Claude, Cursor, the Apify MCP server, and similar). The agent calls a tool, gets structured results back, and continues its workflow — enrichment, qualification, or research — without leaving the conversation.
How to use it
Connect the Sales Intelligence MCP to your MCP client and let the agent call its tools. A sample agent instruction:
Using the Sales Intelligence tools, look up "Acme Robotics" and return its industry, size, and key contacts.
Pricing is transparent Pay-Per-Event at $8 per tool call — each call runs a full intelligence lookup.
How to connect it
Add the server to any MCP-capable client — Claude, Cursor, or the Apify MCP server — and the agent can call its tools immediately. A representative instruction is simply to name the company and the fields you want back; the agent picks the right tool and returns the result inline. Because it runs on Apify, you manage it with the same account, billing, and logs as the rest of your data tooling.
Common workflows
Typical uses are: a lead-qualification agent that looks up an inbound company and returns size, industry, and contacts before routing it; a research assistant that assembles a company brief on demand; an enrichment step in an automated pipeline that fills missing firmographics; and an interactive analyst that a rep can ask about an account in plain language. Because the tools return structured results, the agent can act on them — score, route, summarise — not just display them.
Why MCP instead of a plain API
The Model Context Protocol is the emerging standard for letting AI agents call tools directly. An MCP server advertises its tools and their inputs to any MCP-capable client, so the agent discovers what it can do and calls it natively — no bespoke integration code, no prompt-glue to translate between the model and a REST endpoint. For a builder, that means you connect the server once and the agent can use it in any conversation, rather than hand-wiring each call.
A note on agentic payment
Apify’s experimental x402 support lets agents pay for standard Pay-Per-Event Actors in USDC with no account. MCP servers, however, run in Standby mode, which x402 does not currently cover, so this server is called with a standard Apify API key or through the Apify MCP server rather than via x402. That is a limitation of how Standby actors work today, and it is worth knowing up front if your design assumed account-free agent payment.
Pricing, honestly
Pricing is transparent Pay-Per-Event at $8 per tool call, and each call runs a full intelligence lookup rather than a trivial request — so the cost maps to a completed piece of research, not a page fetch. There is no subscription; you pay for the calls your agent actually makes. For high-volume automations, budget against the number of lookups you expect, and cache results your workflow will reuse rather than re-querying the same company repeatedly.
FAQ
Which clients work? Any MCP-capable client, including the Apify MCP server.
Can it pay via x402? Not this one — x402 covers standard Pay-Per-Event Actors; MCP servers run in Standby mode, which x402 does not cover. Use a standard Apify API key or the Apify MCP server.
Is it for humans or agents? Primarily agents and automation, though the same tools can be called programmatically.
Which clients work? Any MCP-capable client, including Claude, Cursor, and the Apify MCP server.
Is it for humans or agents? Primarily agents and automations, though the same tools can be invoked programmatically by a script.
Can it pay via x402? No — x402 covers standard PPE Actors; this is a Standby MCP server, so use a standard API key or the Apify MCP server.
Connected once and called natively, it turns company and lead research into something your agents simply do, rather than something a person has to stop and look up.
Related: LinkedIn Jobs Scraper, Website Contact Scraper, and Startup Funding Tracker for the data behind sales workflows.
We cover AI agents, automation, and the tools that make them work. Our mission is to make AI accessible to everyone.