datos.gob.do-MCP-server

Run GROUP BY + aggregations against a cached resource without writing SQL. Typed wrapper that builds safe DuckDB queries from JSON. Example usage: "How many employees by status in April 2026?" → aggregations=[{col: null, fn: count, alias: empleados}], group_by=["Estatus"], filters=[{col:"Año",op:"=",val:2026},{col:"Mes",op:"=",val:"Abril"}], order_by=[{col:"empleados",dir:"desc"}]. First call downloads + caches the file. Subsequent calls reuse the cache. Returns one row per group with the aggregation values.

Autocomplete dataset / organization / group / tag names. Resolves a slug when the user gives only part of a name. Slugs here are the full registered name, so acronyms rarely match on their own: kind='organization', query='indotel' → 'instituto-dominicano-de-las-telecomunicaciones-indotel'. Returns `{suggestions, count, kind, query}`; the echo keeps a zero-result reply interpretable.

Ask whether each file can actually be downloaded, without downloading it. The catalog says a resource exists; it does not say the file is still there. Call this before recommending a source — recommending a dataset whose files all refuse sends the user to a dead end, and nothing else here tells you in advance. Returns a class, not a yes/no: ok, challenge (a browser passes, no client can), waf_rule, not_found, server_error, html_page, head_not_supported, network.

Remove all cached Parquet files. Returns the count removed.

Find rows where a numeric column falls outside the IQR fence. Uses the standard IQR method: outliers are values below Q1 - 1.5*IQR or above Q3 + 1.5*IQR. Returns rows sorted by distance from the median. Useful for detecting data-entry errors in salary, budget, or census data. First call downloads + caches. Subsequent calls reuse the cache.

Download a resource and return N rows with their column headers. The datos.gob.do portal has no DataStore (no SQL), so this tool downloads the file and parses it client-side. 5 MB cap to avoid huge files. Useful for inspecting the structure of the data before deciding how to query it. For analytical queries on big files, use get_resource_schema + summarize_resource (v0.2) or aggregate_resource (v0.3+).

Run a typed WHERE/SELECT/ORDER BY/LIMIT against a cached resource. First call downloads the file (up to 100 MB) and caches it as Parquet at ~/.cache/datosgobdo-mcp/. Subsequent calls hit cache (<1s). Returns requested columns + matching rows (capped at limit) plus the total count of matching rows. Use this when you need actual records, not aggregates.

Find rows that appear more than once on the given columns (or all columns). Returns duplicate groups sorted by frequency descending. Useful for detecting data-quality issues in payroll, census, and registry datasets. First call downloads + caches. Subsequent calls reuse the cache.

Return cache stats (entries, bytes) plus server identity: version, security mode.

Full metadata for one dataset, including every downloadable resource. Returns title, description, institution, licence, and the complete resource list (CSV/XLSX/PDF/… files) with direct download URLs.

Details for one institution: description, dataset count, URL.

Metadata for one resource (file): download URL, format, size.

Return column names, inferred types, and sample values for a resource. Cheap reconnaissance step. Downloads file (up to 100 MB), opens it in DuckDB, and runs DESCRIBE + per-column DISTINCT sampling. Does NOT return raw rows. Use this before summarize_resource or aggregate_resource so the model knows column names and types.

Portal-wide totals for datos.gob.do: datasets, institutions, groups, tags.

Thematic categories (economy, health, public administration, …) → `{groups, count}`.

List the government institutions publishing on datos.gob.do. Ministries, autonomous agencies, municipalities and so on, with a dataset count each. Returns `{organizations, count, limit_reached}`: 266 exist and `limit` caps at 200, so a full page is a page, not the set.

Most recently modified datasets on datos.gob.do. For monitoring portal updates. Returns hydrated metadata, not raw activity records.

Tags, optionally filtered by prefix → `{tags, count, limit_reached}`. 874 exist, so a listing without `query` is a sample.

Percentile distribution (p25/p50/p75/p90/p95/p99) of numeric columns. Fills the gap left by aggregate_resource, which only exposes median. First call downloads + caches the file. Subsequent calls reuse the cache. Useful for salary analysis, budget distributions, and statistical profiling.

Run an ad-hoc read-only SQL query against a cached resource via DuckDB. Power-user escape hatch when filter_resource / aggregate_resource don't cover the case. The cached resource is exposed as the in-memory table 'data'. SQL is DuckDB dialect — see https://duckdb.org/docs/sql/introduction. Supports CSV, TSV, XLSX, XLS, JSON, and ODS (auto-converted to CSV). Safety: - Only SELECT/WITH statements (CTEs allowed); multi-statement blocked. - DDL/DML keywords (INSERT/UPDATE/DELETE/DROP/CREATE/ALTER/COPY/EXPORT/ IMPORT/TRUNCATE/GRANT/REVOKE/PRAGMA/SET/LOAD/INSTALL/ATTACH/DETACH/ VACUUM/ANALYZE) rejected outright. - Sandboxed: the resource is materialized in memory and external access is disabled, so table functions (read_text/read_csv/glob/...) cannot read local files or reach the network. - Row cap always applied via outer wrapper.

Write a query or filter result to a local CSV file. Export endpoint for analysis workflows — run your filter or SQL, then save the result to open in Excel or another tool. Returns the file path and row count. First call downloads + caches the source file. Subsequent calls reuse the cache.

Search datasets on datos.gob.do (Dominican Republic open data). Filter by keyword, institution, tag or thematic group. Returns summary metadata: title, publishing institution, available formats, URL.

Search individual files by name. Returns download URLs, with the dataset and institution behind each one.

Auto-generated profile: row count, types, nulls, distinct, min/max/mean, top values. Downloads file (up to 100 MB), runs DuckDB COUNT/DISTINCT/AGG queries per column. Returns one compact dict per column with stats. The model uses this to decide which filters and aggregations to apply next, without any raw rows in its context. For columns with many distinct values (e.g. names), 'top_values' is omitted; only counts are returned.