Server has 4 tools with consistent naming (verb_noun pattern: get_*, list_*) and descriptions present for all tools and parameters. However, input schemas are inferred from code attributes rather than explicitly visible in JSON Schema format. Parameter descriptions exist but are minimal (10-30 chars), below the 50-200 char baseline for LLM-optimized descriptions. Output schemas are not documented in the tool definitions. Error handling returns structured error objects but lacks recovery guidance (e.g., 'try search_users() first'). No tool annotations (readOnlyHint, destructiveHint) despite all being READ_ONLY. Tool descriptions average ~150 chars, which is acceptable, but parameter descriptions are sparse.

• Document output schemas for each tool in the tool description. E.g., 'Returns: {symbol, name, price, change, changePercent, volume, marketCap, currency, timestamp, stockType}'. This enables LLMs to plan chained calls.
• Expand parameter descriptions to 50-100 chars with format guidance. E.g., 'Stock symbol (3-6 chars, uppercase). Use base symbol (JKH) for voting shares, .N0000 suffix for voting, .X0000 for non-voting. Examples: JKH, JKH.N0000, TESS.X0000.'
• Add recovery guidance to error messages. E.g., 'Symbol not found. Call list_company_stocks(ticker="JKH") to see available stock types for this company.'
• Implement tool annotations via [McpServerTool] attributes. Add readOnlyHint=true to all four tools to signal they are safe to retry and do not modify state.
• Consider adding a search_stocks or list_all_stocks tool to enable discovery when users provide partial or incorrect symbols. This reduces failed calls and improves user experience.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry