What is the Model Context Protocol (MCP)?
The Model Context Protocol (MCP) is an open protocol that standardises how AI applications connect to external tools and data. An AI application (the host) runs MCP clients that connect to MCP servers, and each server offers tools, resources and prompts over JSON-RPC 2.0. MCP replaces one-off integrations with a shared interface, but it does not enforce security itself: the specification makes the host responsible for user consent before any tool runs.
- Published
- 22 October 2026
- Sources checked
- 22 October 2026
- From
- Kramiva
01
The parts
- Host: the AI application the user interacts with
- Client: a connector inside the host, one per server
- Server: a program that exposes capabilities to the model
- Tools: functions the model can call, such as querying a database or creating a ticket
- Resources: data the application can read, such as files or records
- Prompts: reusable templates a server offers to users
02
How a tool call works
When a host connects, the server lists what it offers. The model sees the tool descriptions, decides to call one, and the host sends the request through the client to the server, which runs it and returns the result. Servers can also ask the user for more information through the host, a client feature the specification calls elicitation.
03
Security is the host's job
The specification is direct about this: tools represent arbitrary code execution, hosts must obtain explicit user consent before invoking any tool, and tool descriptions from a server should be treated as untrusted. It also states that MCP cannot enforce these principles at the protocol level — implementers have to.
04
Trade-offs
MCP's value is reuse: one server can serve any compatible application. The cost is another moving part, and a well-known debate — one of the most discussed Hacker News threads on the subject in 2026 asked when MCP makes sense compared with simply letting an agent run a command-line tool. For a single application calling a few internal functions, plain function calling may be simpler.
Where Kramiva fits
Kramiva's AI automations connect to the business tools a client already uses. Whatever the connection, irreversible actions — sending, paying, deleting, publishing — require a person's approval.
Sources
- Model Context Protocol specification — Model Context Protocol
- When does MCP make sense vs CLI? (discussion) — Hacker News, 1 March 2026
Accessed 22 October 2026. Kramiva is not affiliated with these publishers.
Related questions
Tell us what you are building.
A short brief gets you a real reply from a founder within one business day — an honest read on fit, scope, and budget.