Back to Glossary Index
Core ConceptModel Input Layer

Model Context

Industry Definition Set • Entity Resolution Path: /glossary/model-context

Quick Answer / TL;DR

The total information available to an LLM in a single request, including the system prompt, conversation history, tool definitions, and retrieved documents.

Key Takeaways

  • Total information available to the LLM in one request.
  • Includes prompt, history, tools, and retrieved documents.
  • Limited by the model's context window.
  • Optimizing context is key to effective MCP agent performance.
Definitive Statement: The total information available to an LLM in a single request, including the system prompt, conversation history, tool definitions, and retrieved documents.

Technical Context & Protocol Usage

Detailed Explanation
Model context is everything the LLM 'sees' when generating a response. In MCP systems, context includes the system prompt, the conversation so far, definitions of all available tools, any retrieved documents (from RAG), and previous tool results. The context window limits how much information can be included. Effective MCP implementations optimize context usage to include only the most relevant information.

Format & Payload Metadata

Format: Combined text and structured data within context window

Latency: Larger contexts increase inference latency and cost

Real-World Implementation Use Case

An MCP client carefully manages context by truncating old messages and only including relevant tool definitions to stay within the 128K token window.

M
MCPserver.in Engineering

Platform Team

Published: 2026-07-20
Updated: 2026-07-20

Cite This Page

MLA Style:

MCPserver.in Engineering. "Model Context." MCPserver.in Knowledge Hub, 20 July 2026, mcpserver.in/glossary/model-context.