Server has 5 tools with basic schemas and descriptions, but significant gaps limit production readiness. All tools have descriptions (10-60 chars, mostly adequate) and input schemas with type definitions. However, output schemas are completely undocumented, LLMs cannot plan downstream calls or extract structured data. Parameter descriptions lack format constraints, ranges, and validation guidance. No error handling guidance; tools return raw exception strings. Tool names follow verb_noun convention (good), but descriptions are terse and lack WHEN/WHY context. No pagination support despite async report tool suggesting large datasets. Missing idempotency guidance for write operations.

• Document output schemas for all tools. Example for get_campaign_spend: 'Returns {"data": {"spendPerCampaign": {"results": [{"campaignId": int, "campaign": string, "spend": float}]}}}'
• Add format constraints to date parameters: 'from_date: string, format YYYY-MM-DD, required. Example: 2024-01-15. Must be <= today.'
• Replace raw exception strings with structured error responses: '{"error": "invalid_date_range", "message": "from_date must be before to_date", "suggestion": "Try from_date=2024-01-01, to_date=2024-01-31"}'
• Add pagination to create_async_report and get_campaign_spend: 'limit: int (1-1000, default 100), offset: int (default 0). Returns total_count and next_offset.'
• Document idempotency for create_async_report: 'Idempotent if called with identical dimensions/metrics within 1 hour. Returns existing report_id if duplicate detected.'