Three well-named, read-only tools with clear descriptions and complete input schemas. 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 type definitions and parameter descriptions. However, output schemas are not documented, LLMs cannot predict what fields to expect from responses. Parameter descriptions 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 actionable error messages. No tool annotations (readOnlyHint, idempotentHint) despite all being read-only and idempotent. Missing documentation of response structure and field mappings needed for chaining (e.g., what fields does search_datasets return that query_layer needs?).

• Document output schemas for all three tools. Example: search_datasets returns {total: number, count: number, datasets: [{id, name, summary, org, org_id, record_count, type, url}]}. query_layer returns ArcGIS query response with features array and metadata. layer_info returns layer schema with fields array, geometry type, and capabilities.
• Formalize parameter constraints in JSON Schema: add minItems/maxItems to limit (1-50 for search_datasets, 1-2000 for query_layer), mark url as required with pattern validation, use enum for out_fields if only specific field sets are valid.
• Add tool annotations: readOnlyHint: true and idempotentHint: true to all three tools. This signals to agents that calls are safe to retry and cache.
• Enhance error messages with recovery guidance. Example: 'ArcGIS Hub returned 404. Verify org_id is correct or try search_datasets with a different query.' For query_layer: 'Invalid where clause. Check SQL syntax or use where="1=1" to sample all records.'
• Add examples to parameter descriptions showing valid URL formats and query patterns. Example for query_layer.where: 'e.g., "ZONING = \'Commercial\'" or "YEAR >= 2020"' (but avoid literal example values that LLMs might reuse).