TypeScript MCP server template — plug-and-play tools (stock quotes, weather, RSS, file), resources & prompts examples
Server has 5 tools with complete schemas and descriptions. Naming follows verb_noun convention (get_*, search_*, read_, count_). Descriptions are present (avg ~120 chars) but lack LLM-optimized guidance on when to use each tool vs. alternatives. Parameters are typed with constraints (minItems, maxItems, minLength, format), but descriptions are minimal (avg ~40 chars) and lack dependency hints. Output schemas are not documented in the source. Error handling is basic, no recovery guidance or actionable error messages visible. Composition is sound: each tool has one responsibility, and batch support exists (get_stock_quote accepts array). No security issues detected (read-only operations, no secrets in params). Prompts and resources are registered, showing good feature coverage.
统计一个本地文本文件的行数、词数和字符数(类似 wc)。仅支持纯文本文件。
查询指定城市的当前天气(温度、湿度、天气现象、风速),数据源 Open-Meteo,无需 API key。
获取股票实时行情,支持 A股、港股、美股,可批量查询。代码格式灵活:A股可传 600519 / sh600519 / sz000001;港股可传 00700 / hk00700;美股传 AAPL / TSLA 等 ticker。若不确定代码,先用 search_stock 按名称搜索。
读取 RSS/Atom 订阅源的最新条目(标题、链接、发布时间)。
按名称、拼音或代码片段搜索股票,返回股票代码及所属市场(A股/港股/美股)。
Output schemas not documented. LLMs cannot infer what fields to expect from tool responses (e.g., Quote object structure, RSS entry fields). This forces agents to guess downstream field names and risks broken tool chains.
Parameter descriptions are too brief (avg 40 chars). E.g., 'symbols' param says 'stock codes list' but does not explain format flexibility (600519 vs sh600519 vs hk00700) or when to use search_stock first. LLMs cannot infer these nuances.
Tool descriptions lack 'when to use' guidance. E.g., get_stock_quote and search_stock both relate to stocks, but the description does not explain: 'Use search_stock if you only have a company name; use get_stock_quote if you have the code.' This forces LLMs to reason about tool selection.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 67 | 2026-07-28+ | v2 |
No error handling guidance in tool descriptions. E.g., count_file_words does not document what happens if the file does not exist, is binary, or is too large. Agents have no recovery path.
Tool descriptions contain example values (e.g., '600519 / sh600519 / sz000001' in get_stock_quote). LLMs tend to reuse examples literally rather than adapt to context, risking incorrect stock codes.