Three tools with complete input schemas and reasonable descriptions, but significant gaps in output documentation and error handling. Tool names follow verb_noun convention (search_, get_, lookup_). Descriptions are present (80-120 chars) but lack context on when to use each tool vs. alternatives. Parameters are well-typed with enums where appropriate (law_type, status, template_type, language variants). However, no documented output schemas, no error recovery guidance, and no indication of pagination behavior for search results. Missing tool annotations (readOnlyHint, idempotentHint). Database schema visible but tool response shapes are not.

• Document output schema for each tool. Example for search_laws: {results: [{law_id: string, title_th: string, title_en: string, year: number, status: string, category: string}], total: number, limit: number, offset: number}
• Add error recovery hints to descriptions. E.g., search_laws: 'If no results, try removing year_from/year_to filters or broadening the query term.'
• Clarify tool selection criteria in descriptions. E.g., 'Use search_laws to find legislation by keyword or category. Use glossary_lookup to understand specific legal terminology.'
• Implement and document pagination in search_laws response: return total_count and next_offset so agents can fetch additional results.
• Add tool annotations to the MCP tool registration: mark all three as readOnlyHint=true and idempotentHint=true to signal safety to agents.