Gateway MCP Pattern

Gateway MCP Pattern

Tool-access architecture where the agent sees one small MCP router while backend service schemas are loaded or invoked lazily. It preserves structured MCP-style outputs without injecting every service’s full tool catalog into every prompt.

Key points

  • In [src-041], the Nexus-Dev gateway exposes a single routing tool with about 20 tokens of fixed schema overhead.
  • The gateway has slightly higher per-operation cost than Native MCP because each call includes a routing envelope, but the fixed cost remains tiny even as backend services grow [src-041].
  • For a 20-prompt session with two GitHub operations, the article estimates the gateway at 892 tokens versus 61,654 for Native GitHub MCP [src-041].
  • Gateway MCP is recommended for services used regularly but not constantly, roughly the 15-40 percent G/N band, and can also compete with on-demand skills in the 5-15 percent band [src-041].
  • The pattern is especially useful for general-purpose coding agents that may need many external services, but only a few in any single session [src-041].

Related entities

Related concepts

Source references

  • [src-041] Marco Mornati — “The Future of Agentic Tooling: MCP Servers vs. CLI A Data-Driven Comparison” (2026-04-27)

Robin Cartier perspective

This page is part of Robin Cartier's working AI knowledge graph: a practical research layer for production AI, recommendation systems, experimentation, GEO, and agentic web readiness.

The useful next step is to connect this concept back to applied product leadership and operating models.

Recommended next

Keep reading from this thread

From 491 indexed pages and articles.

  1. Wiki concept Progressive Context Loading (Skills) The three-level loading pattern Claude Code uses to keep skills lightweight. Level 1: initial search only reads the YAML frontmatter (name + Related by loaded
  2. Wiki concept Agentic Tooling Gives AI agents access to external systems through APIs, MCP servers, CLIs, browsers, and workflow runtimes. Related by gateway
  3. Insight Generative Engine Optimization for AI Search A practical GEO guide for becoming visible in AI-generated answers through machine-scannable content, authority, schema, and monitoring Readers have engaged with this next