{"report":{"v":1,"package":"@cyanheads/eurostat-mcp-server","version":"0.11.4","ranAt":"2026-11-16T11:46:14.115Z","install":{"ok":true,"integrity":"sha512-n5lBGbxpVRP3phEH9MGNP7979FUkFMQVXbpyqfQgxQIP8AWyUwr5DvITcQO+eCN6ReCFuojIGrR8skrwSwnSMQ==","deps":47,"scripts":[],"bins":["eurostat-mcp-server"],"license":"Apache-2.0","engines":"node >=24.0.0","repository":"git+https://github.com/cyanheads/eurostat-mcp-server.git","bin":"eurostat-mcp-server"},"start":{"ok":true,"ms":365,"serverName":"eurostat-mcp-server","serverVersion":"0.11.4","protocol":"2025-06-18","stderr":"59,\"operation\":\"ToolRegistry.registerTool\",\"requestId\":\"WV43P-R3QQX\",\"timestamp\":\"2026-11-16T11:46:24.156Z\",\"toolName\":\"eurostat_download_dataset\",\"msg\":\"Tool 'eurostat_download_dataset' registered successfully.\"}\n{\"level\":20,\"time\":1794829584157,\"env\":\"development\",\"version\":\"0.11.4\",\"pid\":59,\"operation\":\"ToolRegistry.registerTool\",\"requestId\":\"OTV2P-OMZJ1\",\"timestamp\":\"2026-11-16T11:46:24.157Z\",\"toolName\":\"eurostat_dataframe_describe\",\"msg\":\"Registering tool: 'eurostat_dataframe_describe'\"}\n{\"level\":20,\"time\":1794829584157,\"env\":\"development\",\"version\":\"0.11.4\",\"pid\":59,\"operation\":\"ToolRegistry.registerTool\",\"requestId\":\"OTV2P-OMZJ1\",\"timestamp\":\"2026-11-16T11:46:24.157Z\",\"toolName\":\"eurostat_dataframe_describe\",\"msg\":\"Tool 'eurostat_dataframe_describe' registered successfully.\"}\n{\"level\":20,\"time\":1794829584157,\"env\":\"development\",\"version\":\"0.11.4\",\"pid\":59,\"operation\":\"ToolRegistry.registerTool\",\"requestId\":\"GLKIS-XKAIZ\",\"timestamp\":\"2026-11-16T11:46:24.157Z\",\"toolName\":\"eurostat_dataframe_query\",\"msg\":\"Registering tool: 'eurostat_dataframe_query'\"}\n{\"level\":20,\"time\":1794829584158,\"env\":\"development\",\"version\":\"0.11.4\",\"pid\":59,\"operation\":\"ToolRegistry.registerTool\",\"requestId\":\"GLKIS-XKAIZ\",\"timestamp\":\"2026-11-16T11:46:24.157Z\",\"toolName\":\"eurostat_dataframe_query\",\"msg\":\"Tool 'eurostat_dataframe_query' registered successfully.\"}\n{\"level\":20,\"time\":1794829584158,\"env\":\"development\",\"version\":\"0.11.4\",\"pid\":59,\"operation\":\"createMcpServerInstance\",\"requestId\":\"TQXLK-ZFT55\",\"timestamp\":\"2026-11-16T11:46:24.151Z\",\"msg\":\"All MCP capabilities registered successfully\"}\n"},"tools":[{"name":"eurostat_search_datasets","description":"Search the Eurostat catalogue by keyword for dataset codes, with labels, period coverage and theme breadcrumbs; datasets and predefined tables only, no folders. Pass a code to eurostat_get_dataset_info, then eurostat_query_dataset for a slice or eurostat_download_dataset for the whole dataset. The catalogue joins the dissemination table of contents with the Comext host's dataflow list, which adds the DS-* collections (detailed trade by CN8, HS, SITC, BEC and CPA, and PRODCOM); Comext entries report a last-update date but no period coverage or observation count. A collection on neither list, such as the legacy PRODCOM DS-056120, is not disseminated and cannot be reached.","params":["query","limit","cursor"]},{"name":"eurostat_browse_themes","description":"Navigate the Eurostat theme tree, for discovery when you know the domain but not the dataset code; eurostat_search_datasets finds datasets by keyword. Without theme_code it returns the top-level theme folders, with one that folder's immediate subfolders and datasets. The tree is the dissemination table of contents plus the Comext host's DS-* collections: detailed trade under ext_go_detail (in ext_go) and PRODCOM under prom (in icts). A collection on neither, such as the legacy PRODCOM DS-056120, is not disseminated and appears nowhere in the tree.","params":["theme_code"]},{"name":"eurostat_get_dataset_info","description":"Fetch a Eurostat dataset's metadata: dimensions with sample values, time range, observation count and last update. Call it before eurostat_query_dataset or eurostat_download_dataset to learn the dimension codes their filters take; the download rejects a filter naming a dimension absent here. Up to 10 sample values per dimension; eurostat_get_dimension_values lists the rest. A DS-* code (detailed trade and PRODCOM, any case) is read from the Comext host, which reports no period coverage or observation count; the first call on a large Comext collection downloads its full structure and takes longer, and repeat calls within the hour reuse it.","params":["dataset_code"]},{"name":"eurostat_get_dimension_values","description":"List every value a Eurostat dataset carries in one dimension, with codes and labels exactly as published: when eurostat_get_dataset_info's sample is not enough, or to confirm filter codes before querying. Narrow a long list with filter (words in a code or label) or code_prefix (the codes under a parent, such as one country's regions), and page with cursor. Pass canvas_id to also stage the whole list as a table, to label a download's codes through a join.","params":["dataset_code","dimension","geo_level","filter","code_prefix","limit","cursor","canvas_id"]},{"name":"eurostat_query_dataset","description":"Fetch a filtered slice of a Eurostat dataset as decoded observations: dimension codes and labels, numeric value, OBS_FLAG status and CONF_STATUS confidentiality marker. Returns a deterministic inline prefix (preview_limit) beside totals over the whole match; filters and period inputs shrink the match itself. Read dimension codes with eurostat_get_dataset_info and their values with eurostat_get_dimension_values first. Filter large slices: Eurostat refuses an oversized query (extraction_too_big) or prepares a large one in the background (async_response). On a deployment with a dataframe canvas, a match above 5,000 observations, or any match with stage=true, is staged as a SQL table (tableName). For a whole dataset, eurostat_download_dataset is the cheaper route.","params":["dataset_code","filters","geo_level","since_period","until_period","last_n_periods","preview_limit","stage","lang","canvas_id"]},{"name":"eurostat_download_dataset","description":"Download a Eurostat dataset through the SDMX bulk TSV endpoint and stage it as a SQL table on the dataframe canvas — the route to a whole dataset, where eurostat_query_dataset is the route to a slice. It is expanded into one row per non-blank TSV cell; absentCellCount says which rows are not observations. Filter keys come from eurostat_get_dataset_info and values from eurostat_get_dimension_values. Two budgets bound every transfer: a decoded-byte budget (50 MiB by default) and a time budget counted from the start of the call (55 seconds by default, inside the 60 seconds an MCP client waits), so a download that reaches the time budget returns after about 55 seconds. Reached after rows have streamed, either budget keeps them as a prefix with budgetExceeded set; reached before the first row, the call fails. Only preview_limit rows come back inline; without a canvas the rest are not kept.","params":["dataset_code","filters","since_period","until_period","preview_limit","canvas_id"]},{"name":"eurostat_dataframe_describe","description":"List the tables on a Eurostat dataframe canvas with row counts and column names and types; call it before eurostat_dataframe_query for the names to write SQL against, and read the columns here rather than assuming them. eurostat_query_dataset tables give each dimension a code column (e.g. \"geo\") and a label column (\"geo_label\"); eurostat_download_dataset tables carry code columns only, plus \"time\". Both write the same measure columns (obs_value, obs_flag, obs_flag_label, conf_status, conf_status_label) with the same codes, so they join on dimension codes and time; DS-* tables add obs_value_text for a value published as text. A eurostat_get_dimension_values table has code and label, plus level for geo (NULL for a code with no NUTS level), and labels a download's codes through a join (d.geo = g.code); staged without geo_level, a geo list holds every level, so filter level before aggregating.","params":["canvas_id"]},{"name":"eurostat_dataframe_query","description":"Run one read-only SQL SELECT against the tables staged on a Eurostat dataframe canvas, to reach observations past an inline preview and to aggregate or join them without re-fetching. Call eurostat_dataframe_describe first for the table and column names. Statement chaining, other verbs, and functions that read files or external data are rejected. Returns every complete row up to the canvas row limit.","params":["canvas_id","sql"]}],"toolsHash":"876f5c8557ba39b1392bf07fe1a0f74085a2278c9b85ae46879b1582988bae6e","wrote":[],"host":{"commit":"eaaffffaf52255d1b77dc4aacffdfaf353f0c12c","run":"https://github.com/itsHaddad/bragproof/actions/runs/43857070216"},"network":{"names":["registry.npmjs.org"],"conns":[]}},"summary":{"outcome":"started","needs":[],"contacts":[],"flags":[],"tools":8,"wrote":0,"callsTried":0,"callsAnswered":0,"callContacts":[]},"history":[{"package":"@cyanheads/eurostat-mcp-server","version":"0.11.4","ranAt":"2026-11-16T11:46:14.115Z","outcome":"started","tools":8,"contacts":0,"flags":0,"serious":0,"toolsHash":"876f5c8557ba39b1392bf07fe1a0f74085a2278c9b85ae46879b1582988bae6e"},{"package":"@cyanheads/eurostat-mcp-server","version":"0.11.3","ranAt":"2026-11-12T18:01:28.737Z","outcome":"started","tools":8,"contacts":0,"flags":0,"serious":0,"toolsHash":"4628af9172d4ff305f82d6e7146948843d05e2793b5ad058f273fd01857e162d"}],"page":"https://bragproof.smaverk.com/r/%40cyanheads/eurostat-mcp-server"}