OpenRouter compares context-token costs of MCP, CLI, and Skills
OpenRouter tested the GitHub MCP server, the gh CLI, and a gh skill on one GitHub task across four models, including Claude Sonnet 5.5, Claude Haiku 5.5, GPT-6 Luna, and Gemini 3.8 Flash. On Sonnet 5.5, all three approaches produced correct answers, but input usage differed sharply: 59,417 tokens for the MCP server with 46 default tools, 3,315 for the skill, and 1,421 for the CLI. OpenRouter also reports that its Tool Search reduced the upfront definition cost, though the overall savings varied by model.
The comparison shows that an agent’s tool-access method can materially change context and input-token usage even when the final answer is correct, giving developers a concrete reason to scope tools, trim outputs, or defer tool definitions.
MCP vs CLI vs Skills: What Each Costs in Context
OpenRouter ·11/11/2026


Every tool you give an agent sends its definition to the model on every request, and you pay for it as input tokens. A Model Context Protocol (MCP) server can add tens of thousands of tokens to the context window. The usual alternatives are a command-line tool (CLI) or a skill.
We measured it on one GitHub task with the GitHub MCP server, the gh CLI, and a gh skill, on Claude Haiku 5.5, Claude Sonnet 5.5, GPT-6 Luna, and Gemini 3.8 Flash. On Sonnet 5.5, all three answered correctly, but the input tokens they used were very different. The CLI used 1,421, the skill used 3,315, and the MCP server used 59,417 with its default tools loaded. Here’s where the tokens go and how to cut them with our Tool Search server tool.
#OpenRouter#OpenRouter Tool Search#GitHub MCP server#gh CLI#Claude Sonnet 5.5#Claude Haiku 5.5#GPT-6 Luna#Gemini 3.8 Flash