Tiingo MCP has well-structured tool definitions with complete input schemas and reasonable descriptions. All 4 tools are explicitly registered with name, description, and inputSchema. However, descriptions lack LLM-optimization guidance (e.g., no 'when to use' context), output schemas are not documented, and error handling is minimal. Parameter descriptions are adequate but could be more prescriptive about constraints. The _apiKey parameter pattern is a security anti-pattern, credentials should never be tool parameters.
Get crypto prices for a pair (open, high, low, close, volume) at a chosen resample frequency. Example: crypto_prices({ ticker: "btcusd", resampleFreq: "1day" })
Get recent financial news articles, optionally filtered by tickers or tags. Returns title, description, URL, published date, source, and related tickers. Example: news({ tickers: "aapl,msft", limit: 10 })
Get metadata for a stock ticker: company name, description, exchange code, and the date range of available EOD price data. Example: stock_metadata({ ticker: "AAPL" })
Get EOD historical stock prices for a ticker (open, high, low, close, volume, adjClose). Without a start_date, returns just the latest trading day. Example: stock_prices({ ticker: "AAPL", start_date: "2024-01-01", end_date: "2024-01-31", frequency: "daily" })
API key exposed as tool parameter (_apiKey). Credentials must never appear in tool parameters, they are logged in traces and prompt history. Use server-side secret injection via environment variables.
Output schemas not documented. LLMs cannot plan downstream calls or extract fields without knowing the response structure. Document what fields each tool returns (e.g., stock_prices returns {ticker, prices: [{date, open, high, low, close, volume, adjClose}]})
Error handling is minimal. tiingoGet() returns {error, message} on HTTP failure, but callTool() does not validate required parameters or provide recovery guidance. E.g., if apiKey is missing, return 'No Tiingo key available, configure via environment variable TIINGO_API_KEY' instead of generic error.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 63 | 2026-07-28+ | v2 |
Descriptions lack 'when to use' context. E.g., stock_prices says what it does but not when to call it vs stock_metadata. Add guidance: 'Use this to fetch historical price data for analysis or charting. For current metadata (exchange, date range), call stock_metadata first.'
news tool has no required parameters (all optional). This is valid but risky, calling news() with no args returns 10 articles with no filtering. Add guidance in description: 'At least one of tickers or tags is recommended to narrow results.'