MCP server that dynamically loads and exposes tools from Python scripts in the openapitools directory, with support for Indian stock market APIs
Critical failures across all dimensions. Tool names lack action verbs (IndustrySearchTool, BSEMostActiveStocksRetriever use noun-heavy patterns). Descriptions are minimal (10-50 chars, well below 194-char baseline) and lack context on WHEN to use each tool or what distinguishes similar tools. Input schemas present but severely underdocumented: parameters lack descriptions entirely (e.g., StockForecastsRetriever has 5 required params with zero guidance on valid values, formats, or dependencies). No output schemas documented. No error handling guidance. Tools appear to wrap Python scripts that make external API calls, but the MCP layer provides no recovery paths, retry logic, or actionable error messages. All 16 tools suffer from identical structural deficiencies.
Retrieve the most active stocks from BSE (Bombay Stock Exchange)
Fetch commodities data from the Indian stock market API
Retrieve corporate actions data for a specific stock
Retrieve Indian IPO (Initial Public Offering) data
Retrieve Indian stock market news data
Search for industries by query
Retrieve mutual fund details for a specific stock
Search for mutual funds by query
Tool names lack action verbs. Names like 'IndustrySearchTool', 'BSEMostActiveStocksRetriever', 'CommoditiesDataFetcher' are noun-heavy and do not follow verb_noun convention (get_, list_, search_, fetch_). LLMs cannot infer intent from names alone.
Descriptions are critically short (10-50 chars, baseline 194 chars). Examples: 'Retrieve the most active stocks from BSE' (42 chars), 'Fetch commodities data from the Indian stock market API' (55 chars). No guidance on WHEN to use, what distinguishes similar tools (e.g., BSE vs NSE), or what the output contains.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 25 | 2026-07-28+ | v2 |
Retrieve mutual funds data
Fetch the most active stocks from NSE (National Stock Exchange)
Retrieve price shockers data from the stock market
Fetch recent announcements for a specific stock
Retrieve detailed information for a specific stock
Fetch 52-week high and low data for stocks
Retrieve stock forecasts based on various parameters
Fetch historical data for a stock with specified period and filter
Input parameters lack descriptions entirely. StockForecastsRetriever requires 5 params (stockId, measureCode, periodType, dataType, age) with zero guidance on valid values, formats, enums, or dependencies. LLMs cannot infer what 'measureCode' or 'periodType' accept.
No output schemas documented. Tools return raw JSON from external APIs (e.g., IndustrySearchTool calls stock.indianapi.in and prints response). LLMs cannot plan downstream calls or extract required fields without knowing the response structure.
No error handling or recovery guidance. Python scripts catch requests.exceptions and print JSON errors, but MCP layer does not categorize errors as retryable/user-fixable/fatal or provide actionable next steps. A failed API call returns raw error text with no guidance.