Comparing MCP tool definitions from saved tools/list responses

说明与用途

mcp‑footprint 用于测量 MCP 工具定义的大小，跨服务器比较并生成可视化 HTML 报告。

问题

Comparing large tool definitions across multiple MCP servers is difficult due to raw JSON size and repetition.

结果

mcp‑footprint provides repeatable, server‑agnostic measurements of MCP tool definitions, enabling easier catalog review and comparison across servers.

工作流

• Save complete tools/list JSON or JSON‑RPC responses from each server.
• Run mcp‑footprint CLI with the saved files to generate footprint.html and footprint.json.
• Review the HTML report to see definition byte counts, proportional maps, and searchable rows.
• Use the report to identify exact definitions, same‑name differences, and similar descriptions across servers.
• Iterate after schema changes to compare measurements.
• Optionally use the CLI as a JavaScript library in custom tooling.

证据

I built mcp-footprint to see how large tool definitions are in a captured MCP catalog and which ones repeat across servers.

来源

The CLI sorts object keys recursively before serialization. It counts the UTF-8 bytes in each complete canonical tool definition, then counts the canonical aggregate catalog.

来源

The estimate is ceil(catalogBytes / 4). It's a rough byte-based proxy, not tokenizer output or a measurement of client context use.

来源