io.github.cyanheads/imf-mcp-server

Est-ce que io.github.cyanheads/imf-mcp-server est sûr ?

Au dernier audit MCP Hub, le serveur MCP « io.github.cyanheads/imf-mcp-server » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.

Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.

Les 5 outils

• imf_dataframe_describe≈ 793 tokens

  List DataCanvas tables and columns staged by a prior imf_query_dataset call. Returns each table's name, row count, and column schema (name + DuckDB type). Required before imf_dataframe_query to discover the table and column names for SQL.

  Paramètres, retour & exemple (1)

  Paramètres

  • canvas_id · string · requis
    Canvas ID returned by imf_query_dataset whenever staged=true, from automatic spillover or output_mode="canvas".

  Exemple d'appel

  {"name": "imf_dataframe_describe", "arguments": {"canvas_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"error": {}, "tables": [], "canvas_id": "texte", "table_count": 1}}

• imf_dataframe_query≈ 1197 tokens

  Run a read-only SQL SELECT against a DataCanvas table staged by imf_query_dataset. Supports multi-country comparisons, time-series aggregation, and cross-indicator joins. Requires imf_dataframe_describe first to discover table and column names. One SELECT statement per call; a leading WITH … SELECT (CTE) is accepted. DML and DDL are rejected.

  Paramètres, retour & exemple (2)

  Paramètres

  • sql · string · requis
    Read-only SQL SELECT statement — exactly one statement, starting with SELECT or with a WITH … SELECT common table expression. Reference tables by the names returned by imf_dataframe_describe. Example: SELECT time_period, value FROM spilled_abc123 WHERE time_period >= '2010' ORDER BY time_period.
  • canvas_id · string · requis
    Canvas ID returned by imf_query_dataset whenever staged=true. Call imf_dataframe_describe with it before writing SQL.

  Exemple d'appel

  {"name": "imf_dataframe_query", "arguments": {"sql": "texte", "canvas_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"rows": [], "error": {}, "row_count": 1, "truncated": true}}

• imf_get_database≈ 2675 tokens

  Fetch a dataflow's dimension list with a codelist preview for each dimension. Resolves human-readable terms to SDMX codes (e.g. "United States" → USA, "Constant prices" → NGDP_RPCH). Required before imf_query_dataset — SDMX keys are opaque without codelist lookups. Each codelist is capped at the first 50 entries by default, including previews filtered by codelist_filter. Set dimension_id to retrieve one codelist with bounded limit/offset paging after the optional codelist_filter, which keeps codes whose ID or name contains every word given. Set available_only=true to page codes the dataflow actually publishes, with series and time coverage metadata; availability filtering happens before codelist_filter and paging. The imf://database/{dataflow_id} resource provides the same bounded discovery summary. Country codes are ISO 3-letter (USA, GBR, DEU), not ISO 2-letter (US, GB, DE). The key_format field shows the exact dimension order required by imf_query_dataset. Note: codelists enumerate the code universe, not actual coverage — valid codes can still return no_data if the combination has no series in this dataflow.

  Paramètres, retour & exemple (8)

  Paramètres

  • limit · integer · optionnel
    Entries to return from the selected dimension. Valid only with dimension_id; default 50. One call returns at most 200, so a larger limit (up to 50,000) returns 200, echoes requested_limit and applied_limit, and next_offset names the rest.
  • offset · integer · optionnel
    Matching entries to skip in the selected dimension before this page. Valid only with dimension_id; default 0.
  • version · string · optionnel
    Dataflow version, e.g. 9.0.0. Auto-detected from the dataflow list when omitted.
  • agency_id · string · optionnel
    Agency ID that publishes this dataflow, e.g. IMF.RES or IMF.STA, matched case-insensitively. Auto-detected from the dataflow list when omitted.
  • dataflow_id · string · requis
    Dataflow identifier from imf_list_databases, e.g. WEO, BOP, CPI.
  • dimension_id · string · optionnel
    Exact dimension ID from this tool, e.g. INDICATOR. Select one dimension to page beyond its preview.
  • available_only · boolean · optionnel
    Return only codes reported by the dataflow-wide availability constraint. Default false keeps ordinary codelist discovery unchanged.
  • codelist_filter · string · optionnel
    Optional words to search for in each dimension's codelist, split on spaces and commas. A code matches when every word appears, case-insensitively, in its code ID or name; different words may match different fields. Filtering runs before the 50-entry preview or selected-dimension page. Example: "CPI", "GDP constant prices", or "NGDP_RPCH percent" surfaces matching WEO indicator codes.

  Exemple d'appel

  {"name": "imf_get_database", "arguments": {"dataflow_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"name": "texte", "error": {}, "notice": "texte", "source": "texte", "version": "texte", "agency_id": "texte", "truncated": true, "dimensions": [], "key_format": "texte", "dataflow_id": "texte", "description": "texte", "dsd_version": "texte"}}

• imf_list_databases≈ 1351 tokens

  List IMF SDMX dataflows available on the portal. Entry point for every query: imf_get_database and imf_query_dataset both require a dataflow id obtained here. Vintage (historical snapshot) dataflows such as WEO_2025_OCT_VINTAGE are excluded by default; set include_vintages=true to include them. Results are paged — 50 per call by default and at most 200, adjustable with limit and offset — and total_count reports how many dataflows matched. Descriptions are shortened here; imf_get_database returns the full text for a single dataflow.

  Paramètres, retour & exemple (4)

  Paramètres

  • limit · integer · optionnel
    Maximum dataflows to return in this call. Default 50; one call returns at most 200, so a larger limit (up to 1,000) returns 200 and echoes requested_limit and applied_limit. total_count reports how many matched, so a partial page is always recognizable as one.
  • filter · string · optionnel
    Optional words to filter results by, split on spaces and commas. A dataflow matches when every word appears, case-insensitively, somewhere in its ID, name, or full description. Example: "exchange rate" returns ER and related dataflows; "WEO outlook" returns WEO and the regional outlooks built on it.
  • offset · integer · optionnel
    Number of matching dataflows to skip before this page. Combine with limit to page through a broad or unfiltered catalog.
  • include_vintages · boolean · optionnel
    Include vintage (historical snapshot) dataflows such as WEO_2025_OCT_VINTAGE. Default false — vintages are excluded to keep the discovery surface clean.

  Exemple d'appel

  {"name": "imf_list_databases", "arguments": {"limit": 1, "filter": "texte", "offset": 1, "include_vintages": true}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"cap": 1, "error": {}, "shown": 1, "notice": "texte", "offset": 1, "dataflows": [], "truncated": true, "total_count": 1, "applied_limit": 1, "returned_count": 1, "requested_limit": 1}}

• imf_query_dataset≈ 4998 tokens

  Query an IMF SDMX dataflow by dimension key over a time range. Returns observations with time_period, value, and status, plus the unit, scale, and decimals of each series — a key resolving to several series carries one entry per series in series_metadata, since unit and scale differ between them. An index series also carries the reference_period it is based on (e.g. 2010A, 2010=100) exactly as published, on the series and, where the dataflow publishes one per observation, on each observation. Requires imf_get_database first to obtain the correct key_format and valid dimension codes. Country codes are ISO 3-letter (USA, GBR, DEU — not US, GB, DE). Key format: dot-separated codes in DSD keyPosition order (e.g. USA.NGDP_RPCH.A for WEO). Every position must carry a code: use + to combine codes (e.g. USA+GBR.NGDP_RPCH.A) and * to match every code at a position (e.g. *.NGDP_RPCH.A for all countries); * stands alone and cannot join a + list. Codes are matched case-insensitively and checked against each dimension's codelist before the query — an unknown code is rejected with the nearest valid codes. Codelists from imf_get_database enumerate the code universe, not actual coverage — valid codes can still return no_data if the combination has no series. When a + key returns data for some requested series and none for others, missing_series names each empty one with the reason a query for it alone gives (no_data or no_data_in_range). When the extra request that recovers the unit, scale, or decimals a + key suppresses fails, unresolved_attributes names them and how to get them. start_period and end_period must be valid period strings (YYYY, YYYY-SN, YYYY-QN, YYYY-MM, or a calendar-valid YYYY-MM-DD) with start_period no later than end_period; malformed or reversed ranges are rejected. A bound covers the whole period it names, so end_period 2023 includes 2023-M12 and 2023-Q4. last_n_observations keeps each series' last N observations — 1 is each series' latest value, the cheap way to ask many countries for their latest figure. Values are in base units everywhere, staged canvas rows included: scale is the power of ten the IMF publishes a series in (value / 10^scale is the published figure), never a factor still to apply. A response is held to 100,000 serialized characters. With DataCanvas enabled (CANVAS_PROVIDER_TYPE=duckdb), a larger result (multi-country, long time range) spills to it: call imf_dataframe_describe first to inspect staged tables and columns, then imf_dataframe_query for SQL analysis. Without DataCanvas, a larger result returns its earliest observations with truncated=true, and retrieval_guidance names the last period returned and how to narrow the query.

  Paramètres, retour & exemple (9)

  Paramètres

  • key · string · requis
    Dot-separated dimension codes in DSD keyPosition order. Call imf_get_database to get key_format and valid codes first. Use + to combine codes at one position (e.g. USA+GBR.NGDP_RPCH.A). Use * to match every code at a position — *.NGDP_RPCH.A returns the indicator for all countries, and CAN.*.A every indicator for Canada. Every position needs a code or a *; an empty segment (USA..A) is rejected. A * stands alone at its position — combined with other codes (USA+*) it is rejected. Codes are trimmed and matched case-insensitively (usa resolves to USA), empty and repeated + members are dropped (USA++GBR+USA is queried as USA+GBR, and notice says so), and the response echoes the canonical key. A code missing from its dimension's codelist is rejected before the query, naming the nearest valid codes. The portal limits the request URL, so a + list of a few hundred codes is rejected as too long — use * there, or split the codes across calls. Country codes are ISO 3-letter: USA not US, GBR not GB, DEU not DE.
  • version · string · optionnel
    Dataflow version, e.g. 9.0.0. Auto-detected from dataflow list when omitted.
  • agency_id · string · optionnel
    Agency ID, e.g. IMF.RES or IMF.STA, matched case-insensitively. Auto-detected from dataflow list when omitted.
  • canvas_id · string · optionnel
    Existing canvas ID to accumulate results into across multiple queries. This selects the destination only; it does not force staging. Use output_mode="canvas" to stage an under-budget result.
  • end_period · string · optionnel
    End of time range (inclusive). Same formats as start_period, and must not be earlier than it. The bound covers the whole period it names, so end_period 2023 admits 2023-M12 and 2023-Q4. Observations after this period are excluded from the result.
  • dataflow_id · string · requis
    Dataflow identifier from imf_list_databases, e.g. WEO, BOP, CPI. Matched case-insensitively; the response echoes the catalog spelling.
  • output_mode · string · optionnel · auto | canvas
    Result placement. auto returns an under-budget result inline and spills only when needed. canvas explicitly stages the full result, using canvas_id when supplied or allocating a fresh canvas.
  • start_period · string · optionnel
    Start of time range (inclusive). Accepts any of YYYY (annual), YYYY-SN (semi-annual, e.g. 2023-S1), YYYY-QN (quarterly, e.g. 2023-Q1), YYYY-MM (monthly), or a calendar-valid YYYY-MM-DD (daily), whatever the dataflow's frequency. The bound covers the whole period it names, so start_period 2023 admits 2023-M01 and 2023-Q1. Observations before this period are excluded from the result.
  • last_n_observations · integer · optionnel
    Keep only each series' last N observations, an integer from 1 to 1,000,000; 1 returns each series' latest value. Latest is per series: series end at different periods, and a WEO series ends in its projection years (e.g. 2031), so set end_period to stop at a past year. With start_period or end_period, the last N inside that range. A series with fewer than N observations is returned whole. Omit to return every observation.

  Exemple d'appel

  {"name": "imf_query_dataset", "arguments": {"key": "texte", "dataflow_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"key": "texte", "error": {}, "notice": "texte", "source": "texte", "staged": true, "canvas_id": "texte", "truncated": true, "end_period": "texte", "table_name": "texte", "dataflow_id": "texte", "observations": [], "start_period": "texte"}}

C'est votre serveur ?

Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».

Créer un compte / revendiquer →

Source

https://imf.caseyjhand.com/mcp