MarketPulse has 13 well-named tools with clear verb-noun patterns (get_, list_, detect_, analyze_, explain_). All tools have descriptions (avg ~50 chars, baseline 194 chars) that are concise but lack depth on WHEN to use them or dependencies. Input schemas are present and properly typed with descriptions for all parameters. However, output schemas are NOT documented, tools return JSON strings without specifying field structure, forcing LLMs to infer response shape. Error handling is minimal; no recovery guidance or actionable error messages visible. No tool annotations (readOnlyHint, destructiveHint, idempotentHint). All tools are READ_ONLY, which is good for safety but not declared in schema.

• Document output schemas for all tools. Specify field names, types, and meanings. E.g., get_latest_quote returns {symbol: string, price: float, timestamp: ISO8601, change_pct: float, volume: int}.
• Expand tool descriptions to 100-200 chars. Add WHEN to use: 'Get the most recent price for a stock. Use this for real-time quotes; use get_recent_ticks() for historical price movement.'
• Add readOnlyHint: true to all 13 tool schemas to declare safety and enable client optimizations.
• Enhance error messages with recovery guidance. E.g., 'Anomaly ID not found. Call detect_market_anomalies() to list recent anomalies, then retry with a valid anomaly_id.'
• Document pagination: specify max limit (e.g., 100), default (e.g., 20), and whether response includes total_count or next_cursor for chaining.

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

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