What is MCP?

​

Why MCP matters

• No custom code — add a server URL or command, and the tools appear automatically
• Namespaced — each server’s tools are prefixed to avoid collisions
• Safe — MIRA prompts you before executing any tool the first time

​

How it works

┌─────────────┐      JSON-RPC 2.0        ┌─────────────────────┐
│   MIRA NAE  │ ─── list_tools ────────► │   MCP Server        │
│   or RLM    │ ◄── tool definitions ─── │  (stdio or SSE)     │
│             │ ─── call_tool ─────────► │                     │
│             │ ◄── tool result ──────── │                     │
└─────────────┘                          └─────────────────────┘

1. MIRA connects to the server and calls list_tools to discover available tools.
2. Tool definitions (name, description, JSON schema for inputs) are injected into the engine’s context.
3. When the engine decides to use a tool it calls call_tool with validated arguments.
4. The server executes the tool and returns the result as text or structured data.

​

Transport types

​

MCP tool namespacing

my-server/search_web
my-server/read_file
another-server/query_db

​

Next steps