ISTAT (Istituto Nazionale di Statistica) MCP — Italy's national statistics via the public SDMX 2.1 REST web service
Three well-named tools with clear, actionable descriptions (150 - 250 chars each). All tools have complete input schemas with type definitions and parameter descriptions. Naming follows verb_noun pattern (list_, dataflow_, get_). Descriptions include examples and use cases. However, output schemas are not formally documented, responses are inferred from code comments. Error handling is minimal; no recovery guidance or categorization. No tool annotations (readOnlyHint, destructiveHint). Parameters lack explicit constraints (enums, ranges, patterns) despite domain-specific requirements (e.g., period format, key syntax). Missing per-parameter validation rules in descriptions.
Get the structure (Data Structure Definition) of one ISTAT dataset: its ordered dimensions and, for each, the valid codes. Use this to learn how to build the dot-separated SDMX key for get_data. The key positions correspond to the dimensions in order; an empty position is a wildcard. Example: dataflow_structure({ dataflow_id: "101_1015" }).
Pull observations from an ISTAT dataset. `key` is a dot-separated SDMX dimension filter, one position per dimension in the order given by dataflow_structure; leave a position empty to wildcard it. Example: get_data({ dataflow_id: "101_1015", key: "A.IT...", start_period: "2021", end_period: "2023" }) selects annual (A), REF_AREA=IT (Italy), and wildcards the rest. Omit `key` (or pass "") to fetch all series — caution, this can be large. Returns decoded series with their dimension labels and per-period values.
Browse or keyword-search ISTAT datasets (dataflows). Each result has an `id` (the dataflowId you pass to get_data / dataflow_structure) and an English name. ISTAT publishes ~4,800 datasets, so always pass `query` to filter unless you really want the whole list. Example: list_dataflows({ query: "unemployment" }) or list_dataflows({ query: "GDP" }).
Output schemas not formally documented. Responses inferred from code comments only. LLMs cannot plan downstream operations without knowing return field types and structure.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). All three tools are read-only but this is not declared in the schema, forcing LLMs to infer safety from descriptions.
Parameter constraints not formalized. 'key' parameter accepts dot-separated SDMX format but no regex pattern or format constraint is declared. 'limit' and 'max_series' lack min/max bounds. LLMs may pass invalid values.
Error handling absent. No error responses documented. If ISTAT API fails or returns malformed data, callTool() will throw unstructured errors with no recovery guidance for the LLM.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 60 | 2026-07-28+ | v2 |
No pagination support in list_dataflows. Results are capped at 50 by default, but no next_cursor or offset mechanism is offered. If an LLM needs to iterate through 4,800 datasets, it cannot.