Seven rules for MCP tools that an AI assistant cannot misread

说明与用途

七条规则，帮助设计 MCP 服务器，防止语言模型误读健康数据 API。

问题

Language models misinterpret API tool lists, leading to incorrect or misleading answers when interacting with health data APIs.

结果

The MCP server now reliably presents accurate, non‑misleading information to language models, preventing erroneous dosage calculations, misinterpretation of empty fields, and accidental data writes.

工作流

• 1. Reduce tool count: merge similar endpoints into single tools per question type.
• 2. Document empty field meanings in tool descriptions.
• 3. Explicitly forbid unit conversions and arithmetic in descriptions.
• 4. Label numeric fields with source (studied vs. printed).
• 5. Ensure all tools are read‑only and state this in the description.
• 6. Scope tools by OAuth scopes (evidence.read vs. me.read).
• 7. Return concise summaries and external links, not full licensed text.

证据

When has_confirmed_label is false there is no panel, and that means the match could not be confirmed. It does NOT mean the product has no ingredients. Never tell a reader a product contains nothing.

来源

Our descriptions say: each amount keeps its own unit; do not add amounts across units or items; do not convert IU into a mass.

来源

The study tool returns the title, year, participant count, our summary and the PubMed link.

来源