Two well-named, focused tools with clear descriptions and complete input schemas. Both tools follow verb_noun naming (lookup_ip, check_risk) and have descriptions in the 150-200 char range, meeting baselines. Input schemas are properly typed with required fields and parameter descriptions. However, output schemas are not documented, LLMs cannot predict response structure. Error handling is basic (returns {error: string} but lacks recovery guidance). No parameter constraints (enums, ranges) despite accepting free-form IP strings. Missing tool annotations (readOnlyHint, idempotentHint) despite both being read-only, keyless operations.

• Document output schemas in tool definitions. Add a 'outputSchema' field to McpToolDefinition showing the structure returned by mapResult() and checkRisk(). Example: {type: 'object', properties: {ip: {type: 'string'}, location: {type: 'object', properties: {country: {type: 'string'}, ...}}, ...}}.
• Add tool annotations: set readOnlyHint: true and idempotentHint: true on both tools to signal to clients that these are safe to cache and retry.
• Add JSON Schema pattern constraint to the 'ip' parameter: 'pattern': '^([0-9]{1,3}\.){3}[0-9]{1,3}$|^[0-9a-fA-F:]+$' to validate format at schema level, not just in code.
• Enhance error messages with recovery guidance. Replace 'IP not found: X' with 'IP address X not found in the IPQuery database. Verify the address is correct or try a different IP.'
• Add explicit timeout (e.g., 10s) to fetch() calls and return a clear timeout error: 'IPQuery API did not respond within 10 seconds. Please retry in a moment.' This prevents agent hangs.