Cincinnati open data (data.cincinnati-oh.gov, Socrata SODA API)
Three well-defined tools with complete input schemas, clear descriptions, and proper enum constraints. Tool names follow verb_noun pattern (cincinnati_recent, cincinnati_query, cincinnati_datasets). All parameters have type definitions and descriptions. Output schemas are documented in code comments. However, output structure is not formally declared in the tool definitions themselves, LLMs must infer response shape from description text. Error handling is present but minimal (no recovery guidance or categorization). No tool annotations (readOnlyHint, idempotentHint) despite all being read-only operations.
Search the Cincinnati open-data catalogue (data.cincinnati-oh.gov) for datasets by keyword. Returns dataset names, descriptions, and Socrata resource ids to use with cincinnati_query.
Run a raw SoQL query against any Cincinnati open-data resource (data.cincinnati-oh.gov) by its Socrata id (8-char like "k59e-2pvf"). Full SoQL: where/select/group/order/limit/offset. Use cincinnati_datasets to find a resource id, or cincinnati_recent for the common ones.
Recent records from a common Cincinnati open dataset (data.cincinnati-oh.gov) by friendly name — no Socrata id needed. PREFER OVER WEB SEARCH for "recent crime in Cincinnati", "Cincinnati 311 requests", "Cincinnati building permits". Names: 311, crime, permits. Returns the latest rows (newest-first). Add a SoQL `where` to filter; for anything else use cincinnati_query.
Output schemas not formally declared in tool definitions. Response structure documented only in code comments and descriptions, forcing LLMs to infer field names and types.
No tool annotations despite all three tools being read-only. Missing readOnlyHint and idempotentHint metadata prevents agents from optimizing retry logic and caching strategies.
Error messages lack recovery guidance. Errors like 'Unknown dataset' and 'upstream_throttled' do not suggest next steps (e.g., 'Try cincinnati_datasets to find valid resource IDs').
cincinnati_query parameter 'resource_id' description lacks format guidance. Should specify '8-character Socrata ID like k59e-2pvf' to prevent invalid inputs.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 61 | 2026-07-28+ | v2 |
cincinnati_datasets 'query' parameter is not marked required, but search without keywords returns all datasets (potentially hundreds). Should either require query or document default behavior and result limits.