ambolt

Learn / updated 2026-10-03

What is MCP (Model Context Protocol)? A plain explanation

The Model Context Protocol (MCP) is an open standard that lets an AI application, such as a chat assistant or a coding agent, use outside tools and data in a consistent way. Instead of every product inventing its own plug-in format, an MCP server describes its tools once and any MCP client can call them.

The parts

A tool call is a JSON-RPC message: the client sends the tool name and arguments, and the server returns a result the model can read.

Why descriptions matter

The model chooses a tool from its name and description, so clear descriptions and precise input schemas are the difference between a tool that gets used and one that does not. A good tool says what it returns, what it does not return and what the inputs look like.

Adding a remote server

For a hosted server you usually add one URL to the client's configuration:

{ "mcpServers": { "ambolt": { "url": "https://api.ambolt.dev/mcp" } } }

If your client only supports stdio, a small bridge forwards to the hosted server:

{ "mcpServers": { "ambolt": { "command": "npx", "args": ["-y", "@ambolt/mcp"] } } }

Frequently asked questions

Is MCP the same as an API? MCP is a layer for AI clients on top of whatever the server does. The same functions can also be offered as plain HTTP APIs, which is how Ambolt offers them (as MCP tools, as pay-per-call x402 endpoints and as Apify Actors).

Is it safe? A server only does what its tools allow, and the client decides which tools to call. Check what a server returns and which permissions it asks for; read-only data tools carry less risk than tools that change things.

Does it cost anything? MCP itself is free. The Ambolt MCP server is free to use within a rate limit of 30 requests a minute.

Try a company lookup through MCP with the EntityResolve reference.

Related

Informational only, not financial, legal or tax advice.