MCP server for Yahoo Finance providing stock data, financial statements, and market analysis
The server provides 12 tools with mostly adequate naming and descriptions, but exhibits inconsistent schema quality, missing output documentation, and parameter validation guidance. Tool names follow verb_noun conventions well (get_*, generate_*). Descriptions average ~120 chars and are present for all tools, though many lack detail on when to use the tool versus alternatives, prerequisites, or what specific data structure is returned. Input schemas are visible and mostly typed, but lack enums for constrained parameters (e.g., 'period' and 'interval' in get_historical_stock_prices accept only specific values but are documented as strings, not enums). Output schemas are almost entirely undocumented, tools return JSON strings without specifying field structure, making it difficult for LLMs to plan multi-step sequences. Error handling is minimal; most tools will fail silently or return generic 'Couldn't fetch' messages without actionable recovery guidance.
Generate a market sentiment dashboard with real-time index performance, fear/greed indicators.
Generate a portfolio performance tracking report for the specified stocks.
Generate a deep technical analysis report for a stock, including price trends, moving averages, and support/resistance levels.
Get cashflow for a given stock symbol.
Get the current stock price based on stock symbol.
Get dividends for a given stock symbol.
Get earning dates.
Get historical stock prices for a given stock symbol.
No output schemas documented. Tools return JSON strings without specifying field structure (e.g., get_historical_stock_prices returns 'prices["Close"].to_json(orient="index")' but LLM cannot see the structure). LLMs must guess or parse unstructured output, increasing errors and wasting tokens.
Constrained string parameters lack enum declarations. 'period' in get_historical_stock_prices accepts only specific values ("1d", "5d", "1mo", etc.) but is documented as a free-form string. LLMs may hallucinate invalid values. Should use JSON Schema enum.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-21 | F | 8 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 48 | - | v1 |
Get income statement for a given stock symbol.
Get news for a given stock symbol.
Get the stock price for a given stock symbol on a specific date.
Get the stock prices for a given date range for a given stock symbol.
Minimal error handling. Tools fail silently or return generic messages like 'Couldn't fetch {symbol} current price'. No guidance on what the LLM should do next (retry, ask user, try alternative tool). See src/mcp_yahoo_finance/server.py get_current_stock_price().
Parameter descriptions lack format constraints. 'date' parameter in get_stock_price_by_date states 'in YYYY-MM-DD format' but provides no example or validation guidance. LLMs frequently produce wrong date formats. Should include actionable constraint examples.
Tool descriptions lack context on when to use vs alternatives. Three tools query historical prices (get_stock_price_by_date, get_stock_price_date_range, get_historical_stock_prices) but descriptions do not explain when to prefer one over another. LLM wastes reasoning cycles.