Three well-named, read-only tools with complete input schemas and clear descriptions. All tools follow verb_noun naming (search_*, query_*, layer_info). Descriptions are substantive (100 - 200 chars) and explain what each tool does and when to use it. Input schemas include types and required fields. However, output schemas are not documented, LLMs cannot predict what fields to expect from responses. Parameter descriptions are good but lack explicit constraints (e.g., 'limit: 1-50' is stated in description but not as JSON Schema minItems/maxItems). Error handling is minimal, no recovery guidance or categorization. Security is strong (SSRF guard in layerUrl, no secrets exposed). Overall solid for a read-only GIS tool, but missing output documentation and error guidance.

• Document output schemas for all three tools. Example for search_datasets: 'Returns {total: number, count: number, datasets: [{id, name, summary, org, org_id, record_count, type, url}]}'. This lets LLMs plan downstream calls.
• Add error recovery guidance. Replace 'throw new Error(`ArcGIS Hub: ${res.status}`)' with structured errors: '{error: "dataset_not_found", message: "No datasets match your query. Try broader keywords like \"parcels\" or \"zoning\".", retryable: false}'.
• Enforce parameter constraints in JSON Schema using minItems, maxItems, pattern, or enum. Move 'limit: 1-50' from description into schema minItems/maxItems.
• Cap query_layer results at 50 by default and document pagination. Add a note: 'Results are limited to 50 features. Use offset to paginate. For large datasets, refine your where clause.'
• Add tool annotations (readOnlyHint: true) to signal to MCP clients that these tools are safe to call without confirmation.