Three well-named, read-only tools with clear descriptions and complete input schemas. All tools follow verb_noun naming (search_datasets, query_layer, layer_info). Descriptions are substantive (150 - 200 chars) and explain what each tool does and when to use it. Input schemas include type definitions and parameter descriptions. However, output schemas are not documented, LLMs cannot predict the structure of responses. Parameter descriptions lack explicit constraints (e.g., 'limit: 1-50' is stated in description but not as JSON Schema minumum/maximum). No error handling guidance or recovery paths. Missing tool annotations (readOnlyHint, idempotentHint). SSRF guard in layerUrl() is a security plus, but not exposed in the tool interface.

• Add JSON Schema minumum/maximum/enum constraints to parameters. E.g., 'limit': { type: 'number', minimum: 1, maximum: 50 }. This makes constraints machine-parseable.
• Document output schemas for each tool. E.g., search_datasets returns { total: number, count: number, datasets: Array<{id, name, summary, org, org_id, record_count, type, url}> }.
• Add tool annotations: readOnlyHint: true for all three tools (they only query, never modify). This signals safety to agents.
• Enhance error messages with recovery guidance. E.g., 'ArcGIS Hub returned 404. Verify the org_id is correct or try a different search query.'
• Document query_layer result limits and pagination. E.g., 'Returns up to 2000 features per call. Use offset for pagination. Large result sets may be truncated; consider filtering with where clause.'