Two tools with clear verb-noun naming (search_entities, get_entity) and reasonable descriptions (194 - 200 chars, within baseline 34 - 392). Both have input schemas with types and descriptions. However, output schemas are undocumented, LLMs cannot predict response structure. Parameter descriptions lack format constraints (e.g., 'schema' enum values are listed in description text, not as JSON Schema enum). Error handling is minimal (generic API error messages, no recovery guidance). No pagination support despite search potentially returning many results. Missing tool annotations (readOnlyHint, idempotentHint). Overall: functional but below production baseline for agent-ready tools.

• Document output schema for both tools. Example for search_entities: 'Returns {query: string, total: number, returned: number, entities: [{id, caption, schema, datasets, target, score, names, countries, birth_date, nationality, topics, position, notes}]}'.
• Convert 'schema' parameter to JSON Schema enum: {"type": "string", "enum": ["Person", "Company", "Organization", "Vessel", "Aircraft"], "description": "Entity type filter"}.
• Add pagination to search_entities: accept 'limit' (1 - 100, default 20) and 'offset' (default 0) parameters. Return total count in response so agent knows if more results exist.
• Enhance error messages with recovery guidance. E.g., 'Entity Q7747 not found. Try search_entities() with a partial name or different country filter.'
• Add tool annotations in McpToolDefinition: readOnlyHint: true, idempotentHint: true for both tools to signal safety to agents.