Skip to main content
Glama

OpenArg — Argentina public data

by org.openarg
OverviewDiscussionsAdmin

Obtener datos de una tabla de OpenArg

obtener_datos
Read-onlyIdempotent

Trae filas de una tabla, en CSV, con la fuente oficial.

Para un total, promedio, conteo, mínimo, máximo o ranking NO traigas filas para sumarlas: usá agregar_datos.

  • columnas: las que quieras (por defecto, todas); nombres exactos de describir_tabla.

  • desde / hasta: período, como AAAA, AAAA-MM o AAAA-MM-DD, sobre la columna de fecha (o de año).

  • filtros: hasta 5. Igualdad: {"provincia": "Córdoba"}. Con operador, una lista: [{"columna": "monto", "operador": "mayor_que", "valor": "1000000"}, {"columna": "provincia", "operador": "en", "valores": ["Salta", "Jujuy"]}]. Operadores: =, !=, mayor_que, menor_que, >=, <=, contiene y en. La igualdad y contiene no distinguen mayúsculas ni acentos (en tablas de más de un millón de filas, sí: ahí la respuesta dice en filtros_aplicados qué se buscó tal cual).

  • orden: "asc" (por defecto: primero lo más viejo) o "desc" (primero lo más reciente), por fecha. Para el último dato de una serie, pedí "desc".

  • limite: 1 a 500 filas. Si hay más, la respuesta lo dice y da el offset de la página siguiente (hasta 10.000).

  • columna_fecha: opcional, otra columna de fecha para desde/hasta y el orden. Si ninguna fila cumple los filtros, la respuesta dice qué valores existen. No descuenta preguntas.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
desdeNo
hastaNo
ordenNo
tablaYes
limiteNo
offsetNo
filtrosNo
columnasNo
columna_fechaNo

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changed
    • addedInput schema / properties / columna_fecha
      Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "title": "Columna Fecha"
+}
    • changedInput schema / properties / filtros / anyOf
      Previous value: -[
-  {
-    "additionalProperties": {
-      "type": "string"
-    },
-    "type": "object"
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "additionalProperties": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "integer"
+        },
+        {
+          "type": "number"
+        }
+      ]
+    },
+    "type": "object"
+  },
+  {
+    "items": {
+      "additionalProperties": true,
+      "type": "object"
+    },
+    "type": "array"
+  },
+  {
+    "type": "null"
+  }
+]
    • addedInput schema / properties / offset
      Added value: +{
+  "default": 0,
+  "title": "Offset",
+  "type": "integer"
+}
    • addedInput schema / properties / orden / anyOf
      Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]
    • changedInput schema / properties / orden / default
      Previous value: -"asc"New value: +null
    • removedInput schema / properties / orden / type
      Removed value: -"string"
  2. First observed

TDQS

A4.7/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the readOnly/idempotent annotations: pagination limits (1-500 rows, offset up to 10,000 and next-page offset reported), the case/accent-insensitive filter behavior and its >1M-row exception surfaced via `filtros_aplicados`, and the empty-result behavior ('la respuesta dice qué valores existen'). Missing auth/rate-limit context, but the operational behavior is richly disclosed.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Bullet-structured and front-loaded with the purpose sentence, and each bullet carries distinct semantic content rather than filler. It is dense and long, but almost no sentence is wasted; a slight trim would be possible.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-parameter tool with an output schema present (so return shape needn't be described), the description covers parameter formats, filter grammar, pagination, ordering, edge cases, and sibling routing. An agent has everything needed to call it correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 0%, so the description must carry the full burden and does: it documents `columnas`, `desde`/`hasta` formats (AAAA / AAAA-MM / AAAA-MM-DD), the entire `filtros` grammar (equality object vs. operator list, the seven operators, max 5), `orden` semantics, `limite` range, the pagination `offset`, and `columna_fecha` as an alternative date column. Effectively every parameter is explained.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb+resource ('Trae filas de una tabla, en CSV, con la fuente oficial') and immediately distinguishes its scope from `agregar_datos` for aggregations. An agent can separate this row-fetch tool from its siblings without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit negation: for totals/averages/counts/min/max/rankings, do NOT fetch rows — use `agregar_datos`. It also routes column-name discovery to `describir_tabla`. Clear when-to-use and when-not-to-use guidance with a named alternative.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All OpenArg — Argentina public data toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources