eurostat-mcp-server

Previous value: -"Canvas identifier returned as canvasId by eurostat_query_dataset or eurostat_download_dataset. Identifies the workspace holding the staged tables."New value: +"Canvas identifier returned as canvasId by eurostat_query_dataset, eurostat_download_dataset, or eurostat_get_dimension_values. Identifies the workspace holding the staged tables."

Previous value: -"Canvas identifier returned as canvasId by eurostat_query_dataset or eurostat_download_dataset. Identifies the workspace holding the staged tables."New value: +"Canvas identifier returned as canvasId by eurostat_query_dataset, eurostat_download_dataset, or eurostat_get_dimension_values. Identifies the workspace holding the staged tables."

Previous value: -"Reuse an existing dataframe canvas so this download lands beside earlier results and can be joined against them. Pass the canvasId a previous eurostat_download_dataset or eurostat_query_dataset response returned; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas."New value: +"Reuse an existing dataframe canvas so this download lands beside earlier results and can be joined against them. Pass the canvasId a previous eurostat_download_dataset or eurostat_query_dataset response returned; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas. A download that carries no observations fails as no_results and leaves the canvas untouched."

Previous value: -"Dataset code (e.g., \"nama_10_gdp\"). Required."New value: +"Dataset code (e.g., \"nama_10_gdp\"). Required. A DS-* code — Comext detailed trade or PRODCOM, in any case — is served by the Comext host, which refuses an unfiltered download of its large collections as extraction_too_big: filter it, including freq on the trade flows, which mix annual and monthly series."

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code is not available for dissemination (HTTP 404, SDMX faultcode 100). `invalid_dimension`: A filter names a dimension the dataset does not have, or a value or period range Eurostat rejects (SDMX faultcode 150). `filter_arity`: Eurostat rejected the positional dimension key because it carried the wrong number of positions (SDMX faultcode 140, INVALID_QUERY_NB_FILTERS), meaning the dataset structure has changed since the metadata call. `invalid_period`: since_period or until_period is not a period literal, or names a month, quarter, semester, trimester, week or day that does not exist, or since_period starts after until_period ends. Checked before any request — the bulk endpoint would otherwise roll an out-of-range period into a neighbouring one, and answer an inverted range with the whole series — and SDMX faultcode 140 TIME_PERIOD_FILTER_SPEC_INVALID maps here too. `async_queued`: Eurostat answered HTTP 200 with a SOAP syncResponse queue ticket (status SUBMITTED) instead of data, because the extraction was too costly to serve synchronously. `no_results`: The download completed but carried no populated observation cells. `upstream_fault`: The SDMX endpoint returned a fault this server does not model, or a body that is not a TSV table. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code is not available for dissemination (HTTP 404, SDMX faultcode 100). `invalid_dimension`: A filter names a dimension the dataset does not have, or a value or period range Eurostat rejects (SDMX faultcode 150). `filter_arity`: Eurostat rejected the positional dimension key because it carried the wrong number of positions (SDMX faultcode 140, INVALID_QUERY_NB_FILTERS), meaning the dataset structure has changed since the metadata call. `invalid_period`: since_period or until_period is not a period literal, or names a month, quarter, semester, trimester, week or day that does not exist, or since_period starts after until_period ends. Checked before any request — the bulk endpoint would otherwise roll an out-of-range period into a neighbouring one, and answer an inverted range with the whole series — and SDMX faultcode 140 TIME_PERIOD_FILTER_SPEC_INVALID maps here too. `async_queued`: Eurostat answered HTTP 200 with a SOAP syncResponse queue ticket (status SUBMITTED) instead of data, because the extraction was too costly to serve synchronously. `no_results`: The download completed but carried no populated observation cells. Nothing is staged, and no canvas is created or touched. `extraction_too_big`: Eurostat refused the extraction as too large (SDMX faultcode 413, HTTP 413): past its 5,000,000-row extraction limit, or an unfiltered download of a large DS-* Comext collection, which Eurostat serves only filtered. Covers every dataset, on either host. `upstream_fault`: The SDMX endpoint returned a fault this server does not model, or a body that is not a TSV table. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found",
-  "invalid_dimension",
-  "filter_arity",
-  "invalid_period",
-  "async_queued",
-  "no_results",
-  "upstream_fault",
-  "canvas_not_found"
-]New value: +[
+  "not_found",
+  "invalid_dimension",
+  "filter_arity",
+  "invalid_period",
+  "async_queued",
+  "no_results",
+  "extraction_too_big",
+  "upstream_fault",
+  "canvas_not_found"
+]

Previous value: -"One observation as a flat row: one column per dimension holding its code, \"time\" for the period, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. A label column is null when Eurostat publishes no label for that code."New value: +"One observation as a flat row: one column per dimension holding its code, \"time\" for the period, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. A label column is null when Eurostat publishes no label for that code. Rows of a DS-* dataset also carry obs_value_text after obs_value: a value Eurostat published as text, such as a PRODCOM quantity unit (\"KG\"), kept verbatim with obs_value null; a PRODCOM \":C\" arrives as conf_status \"C\" instead."

Previous value: -"Dataset code (e.g., \"nama_10_gdp\"). Use eurostat_search_datasets or eurostat_browse_themes to find codes."New value: +"Dataset code (e.g., \"nama_10_gdp\", or \"DS-045409\" for a Comext collection). Use eurostat_search_datasets or eurostat_browse_themes to find codes."

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist or is not available for dissemination. `async_response`: Eurostat returned an async response (query too large for the API). Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist or is not available for dissemination. `upstream_fault`: Eurostat returned a dataset structure or content constraint this server cannot read: malformed, truncated, not XML, or missing the requested dataset's dataflow. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found",
-  "async_response"
-]New value: +[
+  "not_found",
+  "upstream_fault"
+]

Added value: +{
+  "description": "Also stage the values on this dataframe canvas as a table with two columns, code and label, holding every value the dimension lists — past the inline cap too — at the geo_level requested, country when it is omitted. Pass the canvasId a eurostat_download_dataset or eurostat_query_dataset response returned, then search the staged table with SQL or join it to a download on the dimension column (e.g., d.geo = g.code). Omit to return the values inline only: this tool never starts a canvas. Ignored on deployments without a dataframe canvas.",
+  "pattern": "^[A-Za-z0-9_-]{10}$",
+  "type": "string"
+}

Previous value: -"Dataset code (e.g., \"nama_10_gdp\")."New value: +"Dataset code (e.g., \"nama_10_gdp\", or \"DS-045409\" for a Comext collection)."

Added value: +{
+  "description": "Dataframe canvas the values were staged on — the canvas_id supplied. Omitted when nothing was staged: canvas_id was omitted, this deployment runs without a dataframe canvas, or the dimension lists no values.",
+  "type": "string"
+}

Added value: +{
+  "description": "The inline cap applied. Present alongside truncated.",
+  "type": "number"
+}

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code or dimension code does not exist. `async_response`: Eurostat returned an asynchronous-response condition. `no_results`: The dataset has no geo values at the effective geo_level. `conflicting_params`: geo_level was combined with a dimension other than \"geo\", where it has no effect. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code or dimension code does not exist. `no_results`: The dataset has no geo values at the effective geo_level. `conflicting_params`: geo_level was combined with a dimension other than \"geo\", where it has no effect. `upstream_fault`: Eurostat returned a dataset structure or content constraint this server cannot read: malformed, truncated, not XML, or missing the requested dataset's dataflow. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found",
-  "async_response",
-  "no_results",
-  "conflicting_params"
-]New value: +[
+  "not_found",
+  "no_results",
+  "conflicting_params",
+  "upstream_fault",
+  "canvas_not_found"
+]

Added value: +{
+  "description": "Where the values past the inline cap are: the staged table when canvas_id was passed, or how to stage them. Present alongside truncated.",
+  "type": "string"
+}

Added value: +{
+  "description": "Values returned inline. Present alongside truncated.",
+  "type": "number"
+}

Added value: +{
+  "description": "Rows written to the canvas table. Equals totalCount. Omitted alongside tableName.",
+  "type": "number"
+}

Added value: +{
+  "description": "Canvas table holding every value as two VARCHAR columns, code and label. Call eurostat_dataframe_describe with canvasId first to confirm it, then join it in eurostat_dataframe_query. Omitted alongside canvasId.",
+  "type": "string"
+}

Previous value: -"Total number of distinct values returned, after geoLevel filtering for geo."New value: +"Number of distinct values the dimension lists, after geoLevel filtering for geo — all of them, including any past the inline cap."

Added value: +{
+  "description": "True when the dimension lists more values than the inline cap and values holds only the first of them. Omitted when values holds every one.",
+  "type": "boolean"
+}

Previous value: -"All dataset-available values for this dimension. For geo, this is the subset at geoLevel."New value: +"Dataset-available values for this dimension, in Eurostat's order — for geo, the subset at geoLevel. Holds every value up to 2,000; past that, the first 2,000, with truncated set and the whole list on the canvas table when canvas_id was passed."

Previous value: -"Dataset code (e.g., \"nama_10_gdp\"). Required."New value: +"Dataset code (e.g., \"nama_10_gdp\"). Required. A DS-* code — Comext detailed trade or PRODCOM, in any case — is served by the Comext host: filter freq on its trade flows, which mix annual and monthly series, and note that product, reporter and partner carry aggregates (TOTAL, EU27_2020) that double-count when summed with their members."

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404 carrying Eurostat error id 100). `no_results`: The query matched no observation cells — including Eurostat HTTP-200 error id 100. The dataset is valid, but no cell carries a value or a status flag for that filter combination and period range. When Eurostat returns the empty table, the error names the filter values that matched nothing (data.unmatchedValues) and the selected periods that carry no value (data.matchedPeriods, the newest 24, with data.matchedPeriodCount counting all of them). `invalid_period`: since_period or until_period is not a period literal, or names a month, quarter, semester, trimester, week or day that does not exist, or since_period starts after until_period ends. Checked before any request; a period Eurostat itself rejects maps here too. `async_response`: Eurostat returned an async warning or HTTP-413 error array — the query matched too many observations. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404 carrying Eurostat error id 100). `no_results`: The query matched no observation cells — including Eurostat HTTP-200 error id 100. The dataset is valid, but no cell carries a value or a status flag for that filter combination and period range. When Eurostat returns the empty table, the error names the filter values that matched nothing (data.unmatchedValues) and the selected periods that carry no value (data.matchedPeriods, the newest 24, with data.matchedPeriodCount counting all of them). `invalid_period`: since_period or until_period is not a period literal, or names a month, quarter, semester, trimester, week or day that does not exist, or since_period starts after until_period ends. Checked before any request; a period Eurostat itself rejects maps here too. `async_response`: Eurostat returned an async warning or an HTTP-413 error array — the query matched too many observations, including an EXTRACTION_TOO_BIG refusal past Eurostat's 5,000,000-row limit. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."

Previous value: -"Numeric observation value, or null when Eurostat reports none — either unavailable in the source data or withheld, in which case confStatus says so."New value: +"Numeric observation value, or null when Eurostat reports none — unavailable in the source data, withheld (confStatus says so), or published as text (valueText holds it)."

Added value: +{
+  "description": "A value Eurostat published as text rather than a number, verbatim — PRODCOM (DS-*) flag and unit indicators such as QNTUNIT publish a unit like \"KG\" — with value null. PRODCOM's \":C\" is decoded to confStatus \"C\" instead. Omitted for numeric and missing values.",
+  "type": "string"
+}

Previous value: -"Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. Call eurostat_dataframe_describe with canvasId first to confirm the table and columns, then eurostat_dataframe_query. Omitted when nothing was staged: either the match was at or below 5,000 observations, or this deployment runs without a dataframe canvas."New value: +"Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. A DS-* table also carries obs_value_text, the valueText of each row. Call eurostat_dataframe_describe with canvasId first to confirm the table and columns, then eurostat_dataframe_query. Omitted when nothing was staged: either the match was at or below 5,000 observations, or this deployment runs without a dataframe canvas."

Previous value: -"Machine-readable failure mode. Declared by this tool: `no_match`: No datasets matched the query string. `invalid_cursor`: The cursor is malformed, came from a different query, or came from a catalogue snapshot that has since refreshed. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `no_match`: No datasets matched the query string — including a query naming a DS-* code that neither the dissemination table of contents nor the Comext dataflow list carries, which is a collection Eurostat does not disseminate. `invalid_cursor`: The cursor is malformed, came from a different query, or came from a catalogue snapshot that has since refreshed. Other values are possible when a failure originates below the handler."

Previous value: -"Dimension filters as a map of dimension code → array of accepted values, applied by Eurostat before the body is sent. Example: {\"unit\": [\"CP_MEUR\"], \"na_item\": [\"B1G\"], \"geo\": [\"DE\", \"FR\"]}. Omit a dimension or pass an empty array to accept every value for it. Do not put \"time\" here — use since_period/until_period. Naming a dimension the dataset does not have is rejected with the dataset's dimension list rather than silently ignored."New value: +"Dimension filters as a map of dimension code → array of accepted values, applied by Eurostat before the body is sent. Example: {\"unit\": [\"CP_MEUR\"], \"na_item\": [\"B1G\"], \"geo\": [\"DE\", \"FR\"]}. Omit a dimension or pass an empty array to accept every value for it. Dimension codes match in any case (\"GEO\" is geo). Do not put \"time\" here — use since_period/until_period. Naming a dimension the dataset does not have is rejected with the dataset's dimension list rather than silently ignored."

Previous value: -"Start of the period range (e.g., \"2020\", \"2023-Q1\", \"2024-01\"), sent as startPeriod. The most effective way to shrink a bulk response: it removes period columns from the TSV rather than blanking their cells."New value: +"Start of the period range, inclusive, sent as startPeriod. Accepted forms: YYYY, YYYY-MM, YYYY-MM-DD, YYYY-Qn (1–4), YYYY-Sn (1–2), YYYY-Tn (1–3), YYYY-Mnn (01–12), YYYY-Wnn (a week the year has, up to 53) or YYYY-Dnnn (a day the year has, up to 366) (e.g., \"2020\", \"2023-Q1\", \"2024-01\"). Extra leading zeros after the letter are dropped (\"2020-W001\" is sent as \"2020-W01\"), a day of the year is sent as three digits (\"2026-D1\" as \"2026-D001\"), YYYY-A1 is sent as YYYY, and a period of another frequency is mapped onto the dataset's own. A malformed or non-existent period (e.g., \"2020-13\") is rejected as invalid_period. The most effective way to shrink a bulk response: it removes period columns from the TSV rather than blanking their cells."

Previous value: -"End of the period range (e.g., \"2024\"), sent as endPeriod. Omit for data through the latest available period."New value: +"End of the period range, inclusive (e.g., \"2024\"), sent as endPeriod, in the same forms as since_period. Omit for data through the latest available period. The range must hold at least one day: a since_period that starts after until_period ends is rejected as invalid_period, while pairs of different frequencies are fine (\"2020-06\" to \"2020\")."

Previous value: -[
-  {
-    "not": {
-      "required": [
-        "error"
-      ]
-    },
-    "required": [
-      "datasetCode",
-      "dimensionsUsed",
-      "rowCount",
-      "missingCount",
-      "periodRange",
-      "bytesRead",
-      "compressed",
-      "budgetExceeded",
-      "observations",
-      "appliedQuery"
-    ]
-  },
-  {
-    "required": [
-      "error"
-    ]
-  }
-]New value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "datasetCode",
+      "dimensionsUsed",
+      "rowCount",
+      "missingCount",
+      "periodRange",
+      "bytesRead",
+      "compressed",
+      "budgetExceeded",
+      "observations",
+      "totalCount",
+      "appliedQuery"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Removed value: -{
-  "description": "The preview_limit applied to inline observations.",
-  "type": "number"
-}

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code is not available for dissemination (HTTP 404, SDMX faultcode 100). `invalid_dimension`: A filter names a dimension the dataset does not have, or a value or period range Eurostat rejects (SDMX faultcode 150). `filter_arity`: Eurostat rejected the positional dimension key because it carried the wrong number of positions (SDMX faultcode 140), meaning the dataset structure has changed since the metadata call. `async_queued`: Eurostat answered HTTP 200 with a SOAP syncResponse queue ticket (status SUBMITTED) instead of data, because the extraction was too costly to serve synchronously. `no_results`: The download completed but carried no populated observation cells. `upstream_fault`: The SDMX endpoint returned a fault this server does not model, or a body that is not a TSV table. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code is not available for dissemination (HTTP 404, SDMX faultcode 100). `invalid_dimension`: A filter names a dimension the dataset does not have, or a value or period range Eurostat rejects (SDMX faultcode 150). `filter_arity`: Eurostat rejected the positional dimension key because it carried the wrong number of positions (SDMX faultcode 140, INVALID_QUERY_NB_FILTERS), meaning the dataset structure has changed since the metadata call. `invalid_period`: since_period or until_period is not a period literal, or names a month, quarter, semester, trimester, week or day that does not exist, or since_period starts after until_period ends. Checked before any request — the bulk endpoint would otherwise roll an out-of-range period into a neighbouring one, and answer an inverted range with the whole series — and SDMX faultcode 140 TIME_PERIOD_FILTER_SPEC_INVALID maps here too. `async_queued`: Eurostat answered HTTP 200 with a SOAP syncResponse queue ticket (status SUBMITTED) instead of data, because the extraction was too costly to serve synchronously. `no_results`: The download completed but carried no populated observation cells. `upstream_fault`: The SDMX endpoint returned a fault this server does not model, or a body that is not a TSV table. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found",
-  "invalid_dimension",
-  "filter_arity",
-  "async_queued",
-  "no_results",
-  "upstream_fault",
-  "canvas_not_found"
-]New value: +[
+  "not_found",
+  "invalid_dimension",
+  "filter_arity",
+  "invalid_period",
+  "async_queued",
+  "no_results",
+  "upstream_fault",
+  "canvas_not_found"
+]

Previous value: -"Guidance for every staged result, including the required eurostat_dataframe_describe then eurostat_dataframe_query sequence, composed with byte-budget, no-canvas, or empty-result disclosure when applicable."New value: +"Guidance on every download: the staged table with the required eurostat_dataframe_describe then eurostat_dataframe_query sequence (or, without a canvas, what was returned inline and what was discarded), preceded by byte-budget disclosure when the budget stopped the transfer and by the inline-preview length when preview_limit returns fewer rows than were downloaded."

Removed value: -{
-  "description": "Observations returned in the inline preview.",
-  "type": "number"
-}

Added value: +{
+  "description": "Observations the download produced — equal to rowCount. The inline observations array holds only the first preview_limit of them; budgetExceeded, not this count, says whether the download is the whole dataset.",
+  "type": "number"
+}

Removed value: -{
-  "description": "True when the inline observation preview omits rows.",
-  "type": "boolean"
-}

Previous value: -"Dimension filters as a map of dimension code → array of valid values. Example: {\"unit\": [\"CP_MEUR\"], \"na_item\": [\"B1GQ\"], \"geo\": [\"DE\", \"FR\"]}. An empty array is treated as no filter for that dimension and is dropped from the request. Do not include \"geo\" here if using geo_level. Invalid dimension values silently return no data — verify with eurostat_get_dimension_values first."New value: +"Dimension filters as a map of dimension code → array of valid values. Example: {\"unit\": [\"CP_MEUR\"], \"na_item\": [\"B1GQ\"], \"geo\": [\"DE\", \"FR\"]}. An empty array is treated as no filter for that dimension and is dropped from the request. Dimension codes match in any case (\"GEO\" is geo). Do not include \"geo\" here if using geo_level. A value that matches nothing contributes no rows rather than an error; the response names it in unmatchedValues, or in the no_results error when nothing matched at all. eurostat_get_dimension_values lists the valid values."

Previous value: -"Return only the N most recent periods. Mutually exclusive with since_period and until_period."New value: +"Return only the N most recent periods. N counts back from the dataset's latest period, not from the latest period published for this slice, so a slice that lags the rest of the dataset can come back empty — raise N, or use until_period ending at a period the slice has published. Mutually exclusive with since_period and until_period."

Previous value: -"Start of time range (e.g., \"2020\", \"2023-Q1\", \"2024-01\"). Mutually exclusive with last_n_periods."New value: +"Start of the time range, inclusive. Accepted forms: YYYY, YYYY-MM, YYYY-MM-DD, YYYY-Qn (1–4), YYYY-Sn (1–2), YYYY-Tn (1–3), YYYY-Mnn (01–12), YYYY-Wnn (a week the year has, up to 53) or YYYY-Dnnn (a day the year has, up to 366) (e.g., \"2020\", \"2023-Q1\", \"2024-01\"). Extra leading zeros after the letter are dropped (\"2020-Q01\" is sent as \"2020-Q1\"), a day of the year is sent as three digits (\"2026-D1\" as \"2026-D001\"), and YYYY-A1 is sent as YYYY. A period of another frequency is mapped onto the dataset's own, so \"2020-01\" works on annual data. A malformed or non-existent period (e.g., \"2020-13\") is rejected as invalid_period. Mutually exclusive with last_n_periods."

Previous value: -"End of time range (e.g., \"2024\"). Omit for data through the latest available period. Mutually exclusive with last_n_periods."New value: +"End of the time range, inclusive (e.g., \"2024\"), in the same forms as since_period. Omit for data through the latest available period. The range must hold at least one day: a since_period that starts after until_period ends is rejected as invalid_period, while pairs of different frequencies are fine (\"2020-06\" to \"2020\"). Mutually exclusive with last_n_periods."

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404 carrying Eurostat error id 100). `no_results`: The query matched no observation cells — including Eurostat HTTP-200 error id 100. The dataset is valid, but no cell carries a value or a status flag for that filter combination and period range. `async_response`: Eurostat returned an async warning or HTTP-413 error array — the query matched too many observations. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404 carrying Eurostat error id 100). `no_results`: The query matched no observation cells — including Eurostat HTTP-200 error id 100. The dataset is valid, but no cell carries a value or a status flag for that filter combination and period range. When Eurostat returns the empty table, the error names the filter values that matched nothing (data.unmatchedValues) and the selected periods that carry no value (data.matchedPeriods, the newest 24, with data.matchedPeriodCount counting all of them). `invalid_period`: since_period or until_period is not a period literal, or names a month, quarter, semester, trimester, week or day that does not exist, or since_period starts after until_period ends. Checked before any request; a period Eurostat itself rejects maps here too. `async_response`: Eurostat returned an async warning or HTTP-413 error array — the query matched too many observations. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found",
-  "no_results",
-  "async_response",
-  "invalid_dimension",
-  "conflicting_params",
-  "canvas_not_found"
-]New value: +[
+  "not_found",
+  "no_results",
+  "invalid_period",
+  "async_response",
+  "invalid_dimension",
+  "conflicting_params",
+  "canvas_not_found"
+]

Previous value: -"Guidance when preview_limit omits matched rows or the match was staged — distinguishes the inline prefix from filters that reduce the match and, when staged, gives the describe-then-query sequence. Omitted when the preview contains the whole match."New value: +"Guidance when a filter value matched nothing, when preview_limit omits matched rows, or when the match was staged — names the unmatched values, distinguishes the inline prefix from filters that reduce the match and, when staged, gives the describe-then-query sequence. Omitted when every filter value matched and the preview contains the whole match."

Added value: +{
+  "additionalProperties": {
+    "items": {
+      "type": "string"
+    },
+    "type": "array"
+  },
+  "description": "Filter values that matched nothing in the dataset, keyed by dimension code and spelled as sent (matching ignores case). The observations cover only the values that did match. Omitted when every filter value matched.",
+  "propertyNames": {
+    "type": "string"
+  },
+  "type": "object"
+}

Removed value: -1

Added value: +"^[A-Za-z0-9_-]{10}$"

Removed value: -1

Added value: +"^[A-Za-z0-9_-]{10}$"

Previous value: -"Reuse an existing dataframe canvas so this download lands beside earlier results and can be joined against them. Pass a canvasId from a previous response; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas."New value: +"Reuse an existing dataframe canvas so this download lands beside earlier results and can be joined against them. Pass the canvasId a previous eurostat_download_dataset or eurostat_query_dataset response returned; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas."

Added value: +"^[A-Za-z0-9_-]{10}$"

Previous value: -"Reuse an existing dataframe canvas, so a result staged by this call lands beside earlier ones and can be joined against them. Pass a canvasId from a previous response; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas and when the match is at or below 5,000 observations."New value: +"Reuse an existing dataframe canvas, so a result staged by this call lands beside earlier ones and can be joined against them. Pass the canvasId a previous eurostat_query_dataset or eurostat_download_dataset response returned; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas and when the match is at or below 5,000 observations."

Added value: +"^[A-Za-z0-9_-]{10}$"

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]

Added value: +[
+  "string",
+  "number",
+  "null"
+]

Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]

Added value: +[
+  "number",
+  "null"
+]

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The provided theme_code does not exist as a folder in the TOC. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The provided theme_code does not exist in the TOC. `not_a_folder`: The provided theme_code identifies a dataset or table entry instead of a folder. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found"
-]New value: +[
+  "not_found",
+  "not_a_folder"
+]

Previous value: -"Guidance when the download was cut short by the byte budget, when nothing could be staged, or when it returned no observations. Omitted otherwise."New value: +"Guidance for every staged result, including the required eurostat_dataframe_describe then eurostat_dataframe_query sequence, composed with byte-budget, no-canvas, or empty-result disclosure when applicable."

Previous value: -"Canvas table holding every downloaded observation. Omitted when this deployment runs without a dataframe canvas, in which case only the inline preview survives the call."New value: +"Canvas table holding every downloaded observation. Call eurostat_dataframe_describe with canvasId first to confirm the table and columns, then eurostat_dataframe_query. Omitted when this deployment runs without a dataframe canvas, in which case only the inline preview survives the call."

Previous value: -"First 10 dimension values for orientation. Use eurostat_get_dimension_values for the full list. Omitted alongside valuesCount when the value set could not be measured."New value: +"First 10 dataset-available values for orientation. Use eurostat_get_dimension_values for the full constrained list. Omitted alongside valuesCount when Eurostat does not supply a measurable value set."

Previous value: -"Number of distinct values in this dimension. For \"time\" this is the dataset's full period count; every other dimension is counted from the most recent period. Omitted when the value set could not be measured — only \"time\" can be, and an omitted count is unknown, not one. Call eurostat_get_dimension_values for that dimension to obtain it."New value: +"Number of dataset-available values in this dimension, taken from the dataset content constraint. For \"time\" this is the full period count. Omitted only when Eurostat does not supply a measurable value set."

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code or dimension code does not exist. `async_response`: Eurostat returned an async warning — the dimension query matched too many observations. `conflicting_params`: geo_level was combined with a dimension other than \"geo\", where it has no effect. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code or dimension code does not exist. `async_response`: Eurostat returned an asynchronous-response condition. `no_results`: The dataset has no geo values at the effective geo_level. `conflicting_params`: geo_level was combined with a dimension other than \"geo\", where it has no effect. Other values are possible when a failure originates below the handler."

Previous value: -[
-  "not_found",
-  "async_response",
-  "conflicting_params"
-]New value: +[
+  "not_found",
+  "async_response",
+  "no_results",
+  "conflicting_params"
+]

Added value: +{
+  "description": "Effective NUTS hierarchy level for the geo value set. Present only for the geo dimension; country is reported when geo_level was omitted.",
+  "enum": [
+    "aggregate",
+    "country",
+    "nuts1",
+    "nuts2",
+    "nuts3"
+  ],
+  "type": "string"
+}

Previous value: -"Total number of distinct values returned."New value: +"Total number of distinct values returned, after geoLevel filtering for geo."

Previous value: -"All valid values for this dimension in the dataset."New value: +"All dataset-available values for this dimension. For geo, this is the subset at geoLevel."

Previous value: -"Reuse an existing dataframe canvas, so a result staged by this call lands beside earlier ones and can be joined against them. Pass a canvasId from a previous response; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas, and when the result fits inline and nothing is staged."New value: +"Reuse an existing dataframe canvas, so a result staged by this call lands beside earlier ones and can be joined against them. Pass a canvasId from a previous response; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas and when the match is at or below 5,000 observations."

Added value: +{
+  "default": 50,
+  "description": "How many matched observations to return inline, from the deterministic start of the JSON-stat cell order. Default 50; maximum 500. This changes only the inline prefix: it does not reduce obsCount, missingObsCount, timeRange, the upstream response, or the rows staged when the match exceeds 5,000. Use filters or period controls to reduce the match itself.",
+  "maximum": 500,
+  "minimum": 1,
+  "type": "integer"
+}

Previous value: -"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404, Eurostat error id 100). `no_results`: The query matched no observation cells — valid dataset, but no cell carries a value or a status flag for that filter combination and period range. `async_response`: Eurostat returned an async warning — the query matched too many observations. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404 carrying Eurostat error id 100). `no_results`: The query matched no observation cells — including Eurostat HTTP-200 error id 100. The dataset is valid, but no cell carries a value or a status flag for that filter combination and period range. `async_response`: Eurostat returned an async warning or HTTP-413 error array — the query matched too many observations. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler."

Previous value: -"Guidance when the result was capped at 5,000 rows — where the rest of the match is, or how to narrow the query. Omitted for uncapped results."New value: +"Guidance when preview_limit omits matched rows or the match was staged — distinguishes the inline prefix from filters that reduce the match and, when staged, gives the describe-then-query sequence. Omitted when the preview contains the whole match."

Previous value: -"Decoded observations, capped at 5,000 rows. Past the cap these are the first 5,000 the response enumerates — the leading combinations of the dataset dimensions, neither a sample nor the most recent periods. When tableName is set, these same rows and every row past the cap are staged on the dataframe canvas; when it is absent, narrow the query with dimension filters or a period range to choose which observations you get, or switch to eurostat_download_dataset when the whole dataset is what is wanted."New value: +"The first preview_limit decoded observations in deterministic JSON-stat cell order — the leading combinations of the dataset dimensions, neither a sample nor necessarily the most recent periods. This prefix is independent of the 5,000-observation staging threshold. When tableName is set, the table holds every matched row; otherwise use filters or a period range to reduce the match itself."

Previous value: -"Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. The five measure columns match the ones eurostat_download_dataset stages, so the two tables join on dimension codes and time and compare like with like. Omitted when nothing was staged: either the result fit under the cap, or this deployment runs without a dataframe canvas."New value: +"Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. Call eurostat_dataframe_describe with canvasId first to confirm the table and columns, then eurostat_dataframe_query. Omitted when nothing was staged: either the match was at or below 5,000 observations, or this deployment runs without a dataframe canvas."

Previous value: -"True when the result exceeded 5,000 observations, so the returned rows are a prefix of the match rather than all of it. When tableName is set, the whole match is on the canvas and reachable with eurostat_dataframe_query; when it is absent, narrowing the query with dimension filters, or downloading the dataset with eurostat_download_dataset, is what brings the rest into reach."New value: +"True only when the match exceeded the 5,000-observation staging threshold. Independent of preview_limit: observations can be a shorter prefix while truncated is false. When tableName is set, call eurostat_dataframe_describe first and then eurostat_dataframe_query; when it is absent, use filters or period controls to reduce the match."

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "items",
+      "parentPath",
+      "itemCount"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `not_found`: The provided theme_code does not exist as a folder in the TOC. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "not_found"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "items",
-  "parentPath",
-  "itemCount"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "canvasId",
+      "expiresAt",
+      "tables"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `canvas_disabled`: This deployment runs without a dataframe canvas, so there is nothing to describe. `canvas_not_found`: The canvas_id is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "canvas_disabled",
+            "canvas_not_found"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "canvasId",
-  "expiresAt",
-  "tables"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "canvasId",
+      "columns",
+      "rows",
+      "rowCount",
+      "truncated"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `canvas_disabled`: This deployment runs without a dataframe canvas, so there is nothing to query. `canvas_not_found`: The canvas_id is unknown or its lifetime has elapsed. `missing_table`: The SQL names a table that is not staged on this canvas, or that has expired. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "canvas_disabled",
+            "canvas_not_found",
+            "missing_table"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "canvasId",
-  "columns",
-  "rows",
-  "rowCount",
-  "truncated"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "datasetCode",
+      "dimensionsUsed",
+      "rowCount",
+      "missingCount",
+      "periodRange",
+      "bytesRead",
+      "compressed",
+      "budgetExceeded",
+      "observations",
+      "appliedQuery"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "description": "The preview_limit applied to inline observations.",
+  "type": "number"
+}

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code is not available for dissemination (HTTP 404, SDMX faultcode 100). `invalid_dimension`: A filter names a dimension the dataset does not have, or a value or period range Eurostat rejects (SDMX faultcode 150). `filter_arity`: Eurostat rejected the positional dimension key because it carried the wrong number of positions (SDMX faultcode 140), meaning the dataset structure has changed since the metadata call. `async_queued`: Eurostat answered HTTP 200 with a SOAP syncResponse queue ticket (status SUBMITTED) instead of data, because the extraction was too costly to serve synchronously. `no_results`: The download completed but carried no populated observation cells. `upstream_fault`: The SDMX endpoint returned a fault this server does not model, or a body that is not a TSV table. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "not_found",
+            "invalid_dimension",
+            "filter_arity",
+            "async_queued",
+            "no_results",
+            "upstream_fault",
+            "canvas_not_found"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Added value: +{
+  "description": "Observations returned in the inline preview.",
+  "type": "number"
+}

Added value: +{
+  "description": "True when the inline observation preview omits rows.",
+  "type": "boolean"
+}

Removed value: -[
-  "datasetCode",
-  "dimensionsUsed",
-  "rowCount",
-  "missingCount",
-  "periodRange",
-  "bytesRead",
-  "compressed",
-  "budgetExceeded",
-  "observations",
-  "appliedQuery"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "code",
+      "label",
+      "dimensions",
+      "timeRange"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist or is not available for dissemination. `async_response`: Eurostat returned an async response (query too large for the API). Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "not_found",
+            "async_response"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "code",
-  "label",
-  "dimensions",
-  "timeRange"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "dimensionCode",
+      "dimensionLabel",
+      "values",
+      "totalCount"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code or dimension code does not exist. `async_response`: Eurostat returned an async warning — the dimension query matched too many observations. `conflicting_params`: geo_level was combined with a dimension other than \"geo\", where it has no effect. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "not_found",
+            "async_response",
+            "conflicting_params"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "dimensionCode",
-  "dimensionLabel",
-  "values",
-  "totalCount"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "datasetCode",
+      "datasetLabel",
+      "dimensionsUsed",
+      "observations",
+      "obsCount",
+      "truncated",
+      "timeRange",
+      "missingObsCount",
+      "appliedFilters"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `not_found`: The dataset code does not exist (HTTP 404, Eurostat error id 100). `no_results`: The query matched no observation cells — valid dataset, but no cell carries a value or a status flag for that filter combination and period range. `async_response`: Eurostat returned an async warning — the query matched too many observations. `invalid_dimension`: A dimension code in filters does not exist in this dataset (HTTP 400, Eurostat error id 150). `conflicting_params`: Mutually exclusive parameters were combined: \"geo\" filter + geo_level, or since_period/until_period + last_n_periods. `canvas_not_found`: A canvas_id was supplied for staging but is unknown or its lifetime has elapsed. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "not_found",
+            "no_results",
+            "async_response",
+            "invalid_dimension",
+            "conflicting_params",
+            "canvas_not_found"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "datasetCode",
-  "datasetLabel",
-  "dimensionsUsed",
-  "observations",
-  "obsCount",
-  "truncated",
-  "timeRange",
-  "missingObsCount",
-  "appliedFilters"
-]

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +false

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "datasets",
+      "query",
+      "totalMatches",
+      "truncated"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]

Added value: +{
+  "additionalProperties": {},
+  "description": "Present when the call failed. Absent on success.",
+  "properties": {
+    "code": {
+      "description": "JSON-RPC error code for this failure.",
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    "data": {
+      "additionalProperties": {},
+      "properties": {
+        "reason": {
+          "description": "Machine-readable failure mode. Declared by this tool: `no_match`: No datasets matched the query string. `invalid_cursor`: The cursor is malformed, came from a different query, or came from a catalogue snapshot that has since refreshed. Other values are possible when a failure originates below the handler.",
+          "examples": [
+            "no_match",
+            "invalid_cursor"
+          ],
+          "type": "string"
+        },
+        "recovery": {
+          "additionalProperties": {},
+          "description": "Actionable next step for the caller.",
+          "properties": {
+            "hint": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "hint"
+          ],
+          "type": "object"
+        },
+        "retryable": {
+          "description": "Whether retrying may succeed.",
+          "type": "boolean"
+        }
+      },
+      "type": "object"
+    },
+    "message": {
+      "description": "Human-readable description of what went wrong.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "message"
+  ],
+  "type": "object"
+}

Removed value: -[
-  "datasets",
-  "query",
-  "totalMatches",
-  "truncated"
-]

Previous value: -"Number of matched observations with null value (missing data points in the source), counted across everything matched rather than only the returned rows."New value: +"Number of matched observations carrying no numeric value, counted across everything matched rather than only the returned rows. Covers both unavailable and withheld cells — a slice can be wholly confidential, so this equalling obsCount does not mean the data is absent."

Previous value: -"A single decoded observation with dimension values, numeric value, and optional status."New value: +"A single decoded observation with dimension values, numeric value, and the optional OBS_FLAG and CONF_STATUS markers."

Added value: +{
+  "additionalProperties": false,
+  "description": "Eurostat CONF_STATUS for this observation — a different codelist from status. Present when Eurostat restricts the cell, which is usually why value is null. Omitted otherwise.",
+  "properties": {
+    "code": {
+      "description": "CONF_STATUS code: \"C\", \"N\", or \"P\".",
+      "type": "string"
+    },
+    "label": {
+      "description": "Confidentiality description (e.g., \"confidential\", \"not for publication\").",
+      "type": "string"
+    }
+  },
+  "required": [
+    "code",
+    "label"
+  ],
+  "type": "object"
+}

Previous value: -"Status flag for this observation. Omitted for normal observations."New value: +"Eurostat OBS_FLAG for this observation. Omitted for unflagged observations, and never carries a confidentiality code — that arrives in confStatus."

Previous value: -"Status flag code (e.g., \"p\", \"e\", \"d\")."New value: +"OBS_FLAG code (e.g., \"p\", \"e\", \"d\")."

Previous value: -"Numeric observation value, or null when missing (flagged as unavailable in the source data)."New value: +"Numeric observation value, or null when Eurostat reports none — either unavailable in the source data or withheld, in which case confStatus says so."

Previous value: -"Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label. Omitted when nothing was staged: either the result fit under the cap, or this deployment runs without a dataframe canvas."New value: +"Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label. The five measure columns match the ones eurostat_download_dataset stages, so the two tables join on dimension codes and time and compare like with like. Omitted when nothing was staged: either the result fit under the cap, or this deployment runs without a dataframe canvas."

Previous value: -"Canvas identifier returned by eurostat_query_dataset as canvasId. Identifies the workspace holding the staged tables."New value: +"Canvas identifier returned as canvasId by eurostat_query_dataset or eurostat_download_dataset. Identifies the workspace holding the staged tables."

Previous value: -"Canvas identifier returned by eurostat_query_dataset as canvasId. Identifies the workspace holding the staged tables."New value: +"Canvas identifier returned as canvasId by eurostat_query_dataset or eurostat_download_dataset. Identifies the workspace holding the staged tables."

Previous value: -"Dimension code (e.g., \"unit\", \"geo\", \"na_item\"). Use these as filter keys in eurostat_query_dataset."New value: +"Dimension code (e.g., \"unit\", \"geo\", \"na_item\"). Use these as filter keys in eurostat_query_dataset and eurostat_download_dataset."

Previous value: -"Dimension value code. Use these as filter values in eurostat_query_dataset."New value: +"Dimension value code. Use these as filter values in eurostat_query_dataset and eurostat_download_dataset."

Previous value: -"Decoded observations, capped at 5,000 rows. Past the cap these are the first 5,000 the response enumerates — the leading combinations of the dataset dimensions, neither a sample nor the most recent periods. When tableName is set, these same rows and every row past the cap are staged on the dataframe canvas; when it is absent, narrow the query with dimension filters or a period range to choose which observations you get."New value: +"Decoded observations, capped at 5,000 rows. Past the cap these are the first 5,000 the response enumerates — the leading combinations of the dataset dimensions, neither a sample nor the most recent periods. When tableName is set, these same rows and every row past the cap are staged on the dataframe canvas; when it is absent, narrow the query with dimension filters or a period range to choose which observations you get, or switch to eurostat_download_dataset when the whole dataset is what is wanted."

Previous value: -"True when the result exceeded 5,000 observations, so the returned rows are a prefix of the match rather than all of it. When tableName is set, the whole match is on the canvas and reachable with eurostat_dataframe_query; when it is absent, narrowing the query with dimension filters is what brings the rest into reach."New value: +"True when the result exceeded 5,000 observations, so the returned rows are a prefix of the match rather than all of it. When tableName is set, the whole match is on the canvas and reachable with eurostat_dataframe_query; when it is absent, narrowing the query with dimension filters, or downloading the dataset with eurostat_download_dataset, is what brings the rest into reach."

Previous value: -"Dataset code (e.g., \"nama_10_gdp\"). Use this in eurostat_get_dataset_info and eurostat_query_dataset."New value: +"Dataset code (e.g., \"nama_10_gdp\"). Use this in eurostat_get_dataset_info, eurostat_query_dataset, and eurostat_download_dataset."

Added value: +{
+  "description": "Reuse an existing dataframe canvas, so a result staged by this call lands beside earlier ones and can be joined against them. Pass a canvasId from a previous response; omit to start a fresh canvas. Ignored on deployments without a dataframe canvas, and when the result fits inline and nothing is staged.",
+  "type": "string"
+}

Added value: +{
+  "description": "Dataframe canvas holding the staged result. Pass to eurostat_dataframe_describe, eurostat_dataframe_query, or a later eurostat_query_dataset call. Omitted when nothing was staged.",
+  "type": "string"
+}

Previous value: -"Guidance when the result was truncated at 5,000 rows. Omitted for normal results."New value: +"Guidance when the result was capped at 5,000 rows — where the rest of the match is, or how to narrow the query. Omitted for uncapped results."

Previous value: -"Decoded observations, capped at 5,000 rows. Past the cap these are the first 5,000 the response enumerates — the leading combinations of the dataset dimensions, neither a sample nor the most recent periods — so narrow the query with dimension filters or a period range to choose which observations you get."New value: +"Decoded observations, capped at 5,000 rows. Past the cap these are the first 5,000 the response enumerates — the leading combinations of the dataset dimensions, neither a sample nor the most recent periods. When tableName is set, these same rows and every row past the cap are staged on the dataframe canvas; when it is absent, narrow the query with dimension filters or a period range to choose which observations you get."

Added value: +{
+  "description": "Rows written to the canvas table. Matches obsCount. Omitted alongside tableName when nothing was staged.",
+  "type": "number"
+}

Added value: +{
+  "description": "Canvas table holding every matched observation in flat form — one code column per dimension plus a \"_label\" companion, then obs_value, obs_flag, obs_flag_label. Omitted when nothing was staged: either the result fit under the cap, or this deployment runs without a dataframe canvas.",
+  "type": "string"
+}

Previous value: -"True when the result exceeded 5,000 observations and was capped. Apply dimension filters to get the full result."New value: +"True when the result exceeded 5,000 observations, so the returned rows are a prefix of the match rather than all of it. When tableName is set, the whole match is on the canvas and reachable with eurostat_dataframe_query; when it is absent, narrowing the query with dimension filters is what brings the rest into reach."

Added value: +{
+  "description": "Breadcrumbs of the other branches that file this same theme_code. Eurostat lists a few folder codes in more than one branch; the items above come from the first one the catalogue lists, which never has fewer children than the branches named here but can list different ones. theme_code takes a bare code, so those branches cannot be addressed directly — browse down to them from the root instead. Omitted when the code has a single placement — the normal case.",
+  "items": {
+    "description": "Breadcrumb from root to one other folder carrying the requested code.",
+    "items": {
+      "type": "string"
+    },
+    "type": "array"
+  },
+  "type": "array"
+}

Previous value: -"First 10 dimension values for orientation. Use eurostat_get_dimension_values for the full list."New value: +"First 10 dimension values for orientation. Use eurostat_get_dimension_values for the full list. Omitted alongside valuesCount when the value set could not be measured."

Previous value: -"Number of distinct values in this dimension. For \"time\" this is the dataset's full period count; every other dimension is counted from the most recent period."New value: +"Number of distinct values in this dimension. For \"time\" this is the dataset's full period count; every other dimension is counted from the most recent period. Omitted when the value set could not be measured — only \"time\" can be, and an omitted count is unknown, not one. Call eurostat_get_dimension_values for that dimension to obtain it."

Previous value: -[
-  "code",
-  "label",
-  "valuesCount",
-  "sampleValues"
-]New value: +[
+  "code",
+  "label"
+]

Previous value: -"Number of observations with null value (missing data points in the source)."New value: +"Number of matched observations with null value (missing data points in the source), counted across everything matched rather than only the returned rows."

Previous value: -"Decoded observations, capped at 5,000 rows. When truncated is true, apply dimension filters to narrow the result."New value: +"Decoded observations, capped at 5,000 rows. Past the cap these are the first 5,000 the response enumerates — the leading combinations of the dataset dimensions, neither a sample nor the most recent periods — so narrow the query with dimension filters or a period range to choose which observations you get."

Previous value: -"Time coverage of the returned observations. Each bound is omitted when neither the observations nor Eurostat report it — an omitted bound is unknown, not empty."New value: +"Time coverage of everything matched — the same set obsCount counts, so it can reach periods absent from observations when truncated is true. Each bound is omitted when neither the match nor Eurostat report it — an omitted bound is unknown, not empty."

Previous value: -"Most recent period in this result. Omitted when the observations carry no time dimension and Eurostat reports no overall period."New value: +"Most recent period matched. Omitted when the match carries no time dimension and Eurostat reports no overall period."

Previous value: -"Earliest period in this result. Omitted when the observations carry no time dimension and Eurostat reports no overall period."New value: +"Earliest period matched. Omitted when the match carries no time dimension and Eurostat reports no overall period."

Previous value: -"Number of distinct values in this dimension for the most recent period."New value: +"Number of distinct values in this dimension. For \"time\" this is the dataset's full period count; every other dimension is counted from the most recent period."

Previous value: -"ISO 8601 timestamp of the most recent data update."New value: +"ISO 8601 timestamp of the most recent data update. Omitted when Eurostat does not report it."

Previous value: -"Total number of observations in the full dataset (all periods)."New value: +"Total number of observations in the full dataset (all periods). Omitted when Eurostat does not report it — an omitted count is unknown, not zero."

Previous value: -"Overall data coverage period for this dataset."New value: +"Overall data coverage period for this dataset. Each bound is omitted when Eurostat does not report it — an omitted bound is unknown, not empty."

Previous value: -"Most recent available period (e.g., \"2024\")."New value: +"Most recent available period (e.g., \"2024\"). Omitted when Eurostat does not report it."

Previous value: -"Earliest available period (e.g., \"1975\")."New value: +"Earliest available period (e.g., \"1975\"). Omitted when Eurostat does not report it."

Removed value: -[
-  "start",
-  "end"
-]

Previous value: -[
-  "code",
-  "label",
-  "dimensions",
-  "timeRange",
-  "obsCount",
-  "lastUpdated"
-]New value: +[
+  "code",
+  "label",
+  "dimensions",
+  "timeRange"
+]

Previous value: -"NUTS hierarchy level filter — only relevant when dimension is \"geo\". Options: \"aggregate\" (EU/EA codes), \"country\" (2-letter codes, default), \"nuts1\" (3-char), \"nuts2\" (4-char), \"nuts3\" (5-char)."New value: +"NUTS hierarchy level filter — applies only when dimension is \"geo\"; passing it with any other dimension is rejected. Options: \"aggregate\" (EU/EA codes), \"country\" (2-letter codes, default), \"nuts1\" (3-char), \"nuts2\" (4-char), \"nuts3\" (5-char)."

Previous value: -"Dimension filters as a map of dimension code → array of valid values. Example: {\"unit\": [\"CP_MEUR\"], \"na_item\": [\"B1GQ\"], \"geo\": [\"DE\", \"FR\"]}. Do not include \"geo\" here if using geo_level. Invalid dimension values silently return no data — verify with eurostat_get_dimension_values first."New value: +"Dimension filters as a map of dimension code → array of valid values. Example: {\"unit\": [\"CP_MEUR\"], \"na_item\": [\"B1GQ\"], \"geo\": [\"DE\", \"FR\"]}. An empty array is treated as no filter for that dimension and is dropped from the request. Do not include \"geo\" here if using geo_level. Invalid dimension values silently return no data — verify with eurostat_get_dimension_values first."

Previous value: -"Dimension filters that were applied."New value: +"Dimension filters actually sent to Eurostat. Empty arrays from the request are dropped and do not appear here."

Previous value: -"Time coverage of the returned observations."New value: +"Time coverage of the returned observations. Each bound is omitted when neither the observations nor Eurostat report it — an omitted bound is unknown, not empty."

Previous value: -"Most recent period in this result."New value: +"Most recent period in this result. Omitted when the observations carry no time dimension and Eurostat reports no overall period."

Previous value: -"Earliest period in this result."New value: +"Earliest period in this result. Omitted when the observations carry no time dimension and Eurostat reports no overall period."

Removed value: -[
-  "start",
-  "end"
-]

Previous value: -"Folder code to expand (e.g., \"econ\", \"reg\"). Omit to list the top-level theme folders."New value: +"Folder code to expand (e.g., \"economy\", \"reg\"). Omit to list the top-level theme folders."

Previous value: -"Breadcrumb from root to the requested theme (e.g., [\"Economy and finance\", \"National accounts\"]). Empty when browsing root."New value: +"Breadcrumb from root to the requested theme (e.g., [\"Database by themes\", \"Economy and finance\"]). Empty when browsing root."

Previous value: -"Opaque pagination cursor from a previous call's nextCursor. Omit for the first page; pass it back to fetch the next page of matches over a stable order."New value: +"Opaque pagination cursor from a previous call's nextCursor. Omit for the first page; pass it back — with the same query — to fetch the next page of matches over a stable order. A cursor is bound to the query that produced it and to the catalogue snapshot in effect at that time, so reusing one with a different query, or after the catalogue refreshes, is rejected rather than silently paging a different result set."

Previous value: -"Search terms. Split on whitespace into tokens; every token must match (AND), case-insensitively, somewhere across the dataset label, theme breadcrumb, or code. Word order does not matter, so \"business demography NUTS 3\" or \"regional economic accounts\" resolve without naming a label verbatim."New value: +"Search terms — at least one non-whitespace token is required. Split on whitespace into tokens; every token must match (AND), case-insensitively, somewhere across the dataset label, theme breadcrumb, or code. Word order does not matter, so \"business demography NUTS 3\" or \"regional economic accounts\" resolve without naming a label verbatim."

Added value: +"\\S"

Previous value: -"Breadcrumb path from root theme to this dataset (e.g., [\"Economy and finance\", \"National accounts\"]). Empty for top-level entries."New value: +"Breadcrumb path from root theme to this dataset (e.g., [\"Database by themes\", \"Economy and finance\"]). Eurostat files some datasets under several branches; this is the first branch that matched the query. Empty for top-level entries."

Previous value: -"Opaque cursor for the next page of matches. Pass it back as cursor. Omitted on the last page."New value: +"Opaque cursor for the next page of matches. Pass it back as cursor with the same query; it stops working once the catalogue refreshes. Omitted on the last page."

Previous value: -"Total datasets matching the query across all pages, before the page limit."New value: +"Total distinct dataset codes matching the query across all pages, before the page limit."

Previous value: -"Folder code to expand (e.g., \"econ\", \"reg\"). Omit to list the 11 top-level theme folders."New value: +"Folder code to expand (e.g., \"econ\", \"reg\"). Omit to list the top-level theme folders."

Previous value: -"Immediate children of the requested theme, or the 11 root themes if theme_code was omitted."New value: +"Immediate children of the requested theme, or the root themes if theme_code was omitted."

Added value: +{
+  "description": "Suggested next action based on these results. Populated when there is a clear follow-up call.",
+  "type": "string"
+}

Added value: +{
+  "description": "Opaque pagination cursor from a previous call's nextCursor. Omit for the first page; pass it back to fetch the next page of matches over a stable order.",
+  "type": "string"
+}

Previous value: -"Maximum number of results to return (1–100). Default is 20."New value: +"Page size — maximum datasets returned per page (1–100). Default is 20. To retrieve matches beyond one page, pass the returned nextCursor back as cursor; the page size is fixed by this first call."

Previous value: -"Search terms — case-insensitive substring match against dataset labels."New value: +"Search terms. Split on whitespace into tokens; every token must match (AND), case-insensitively, somewhere across the dataset label, theme breadcrumb, or code. Word order does not matter, so \"business demography NUTS 3\" or \"regional economic accounts\" resolve without naming a label verbatim."

Previous value: -"Matching datasets, up to the requested limit."New value: +"Matching datasets for the current page, up to the requested limit."

Added value: +{
+  "description": "Opaque cursor for the next page of matches. Pass it back as cursor. Omitted on the last page.",
+  "type": "string"
+}

Added value: +{
+  "description": "Suggested next action based on these results. Populated when there is a clear follow-up call.",
+  "type": "string"
+}

Previous value: -"Total datasets matching the query before the limit."New value: +"Total datasets matching the query across all pages, before the page limit."

Added value: +{
+  "description": "True when more matches remain beyond this page — pass nextCursor as cursor to fetch them.",
+  "type": "boolean"
+}

Previous value: -[
-  "datasets",
-  "query",
-  "totalMatches"
-]New value: +[
+  "datasets",
+  "query",
+  "totalMatches",
+  "truncated"
+]