Three tools with complete schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools have action verbs (search_, dataset_, query) and non-empty descriptions (avg 156 chars). However, parameter descriptions lack detail on constraints, formats, and dependencies. Output schemas are undocumented, LLMs cannot plan downstream calls or extract structured data. Error handling is minimal. No tool annotations (readOnlyHint, idempotentHint). Composition is sound: search→info→query workflow is logical and tool outputs chain correctly (dataset_id flows through). Security is adequate for read-only operations but lacks explicit permission declarations.

• Document output schemas for all three tools. Example: search_datasets returns {datasets: [{dataset_id, title, theme, record_count}], total_count}. This enables LLMs to extract dataset_id and chain to dataset_info.
• Add ODSQL syntax examples and constraints to query parameters. E.g., 'where: ODSQL WHERE clause (e.g., year >= 2020 AND city = "Paris"). Use field names from dataset_info. Do not use quotes around field names.'
• Add tool annotations to the McpToolDefinition interface: readOnlyHint: true for all three tools, idempotentHint: true for all three. This signals to agents that these are safe to retry.
• Enhance error messages with recovery guidance. E.g., if dataset_id is not found, catch and return: 'Dataset not found. Call search_datasets() with a keyword to find available datasets.'
• Clarify parameter dependencies in descriptions. E.g., 'where overrides query; if both are provided, where takes precedence. Use ODSQL syntax; see dataset_info for available field names.'