Three well-defined tools with complete schemas, clear descriptions, and proper parameter typing. Tool names follow verb_noun convention (kcmo_recent, kcmo_query, kcmo_datasets). All tools have descriptions (194 - 250 chars, within baseline 34 - 392 range). Input schemas are properly typed with required fields declared. However, output schemas are not documented, LLMs cannot predict response structure. Parameter descriptions are adequate but lack constraint details (e.g., 'limit 1-1000' stated in description but not in schema min/max). No error handling guidance or recovery paths documented. No tool annotations (readOnlyHint, etc.) despite all being read-only operations.

• Add JSON Schema outputSchema to each tool definition documenting response structure. Example for kcmo_recent: {type: 'object', properties: {dataset: {type: 'string'}, label: {type: 'string'}, resource_id: {type: 'string'}, rows: {type: 'array', items: {type: 'object'}}, count: {type: 'number'}}}
• Add readOnlyHint: true to all three tools' definitions to signal they are safe to retry and do not modify state.
• Replace '_apiKey' parameter with server-side environment variable injection. Remove from tool schema. Document in README how to configure SOCRATA_API_KEY.
• Add min/max constraints to schema for numeric parameters: limit {type: 'number', minimum: 1, maximum: 1000} instead of relying on description text.
• Enhance error messages with recovery guidance. E.g., 'upstream_throttled: data.kcmo.org rate limit (HTTP 429). Pass _apiKey (Socrata app token) for higher limits.' → 'Rate limited by data.kcmo.org. Retry in 60s or configure SOCRATA_API_KEY environment variable for higher limits.'