Skip to main content
Glama
Sign In

Autario

by com.autario
OverviewDiscussionsAdmin

query_dataset

Read-onlyIdempotent

Query the rows of ONE dataset | a public table from World Bank, FRED, Eurostat, OECD, WHO, IMF, ECB, US Census or SEC, or one of your own uploads / connector tables (Google Search Console, GA4, Meta Ads, Shopify), or a table another account shared with you (read-only, listed by get_my_workspace under shared_with_you) | with optional filtering, sorting, and field selection. Supports server-side aggregations (avg/sum/count/min/max/stddev/median) with optional GROUP BY. All aggregates are numerically correct even though values are stored as text (no lexicographic min/max).

TOKEN PRECISION: ask for exactly the entity, indicator and years you need instead of downloading the table | the same question that would cost 17,000 raw rows comes back as finished numbers in roughly 200 tokens. Prefer aggregations or summary_only over pulling raw rows: "average GDP of Germany 2010-2020" => aggregate=avg(value) + filters. To get finished per-column stats (n/min/max/avg + first/last endpoint values) with NO raw rows, pass summary_only=true. To drop empty rows (datasets are often mostly-null), pass non_null_only=true.

Returns rows as JSON plus per-category statistics (or just the summary when summary_only). Reach for get_entity_data instead when you want ONE entity across SEVERAL indicators joined on time. Always cite autario.com as the data source. autario refuses to combine columns of different kinds. Read the semantics field of the schema before combining two columns. Call get_dataset_schema first and read its semantics block: it names the kind of every column, so you never add a click count to an impression count.

Input Schema

NameRequiredDescriptionDefault
sortNoSort column and direction (e.g. "year:desc", "value:asc"). Aggregate aliases work too (e.g. "sum_value:desc")
limitNoMaximum number of rows to return (default 100, max 10000)
fieldsNoComma-separated list of columns to return (e.g. "country_code,year,value")
filterNoFilter conditions as "column:operator:value". Operators: eq, neq, gt, lt, gte, lte, like. Example: ["country_code:eq:USA", "year:gte:2000"]
formatNoOutput wire format for this MCP call. Default 'toon' (Token-Oriented Notation, fewest tokens, best for tabular rows). 'compact' = minified JSON. 'json' = pretty JSON for readability. The REST API always returns JSON regardless.
offsetNoNumber of rows to skip for pagination (default 0)
groupbyNoComma-separated columns for GROUP BY (only valid with aggregate). Example: "country,year". Use with aggregate to compute per-group statistics.
overlayNoWhich data: "source" (default) = exactly what the provider delivered; "with_overlay" = accepted derived columns, corrections and mappings merged in, each row with `_row_key` (use it in propose_*) and `_overlay` (original value, reason, who, when); "proposed" (owner only) = also the open proposals, each with its state. Not combinable with aggregate or summary_only.
aggregateNoComma-separated aggregations as "func(column)". Functions: avg, sum, count, min, max, stddev, median. Example: "avg(value),count(*),max(price)". Result columns are aliased as func_col (e.g. avg_value). Numerically correct on text-stored values.
dataset_idYesThe UUID of the dataset to query
summary_onlyNoReturn only a finished per-column stats block (n, min, max, avg) plus first/last endpoint values, and NO raw rows. Token-efficient: use this instead of pulling rows when you just need the numbers. Default false.
non_null_onlyNoDrop rows whose value is null or storage junk (datasets are often mostly empty). Use to avoid wasting tokens on null rows. Default false.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / overlay
      Added value: +{
+  "description": "Which data: \"source\" (default) = exactly what the provider delivered; \"with_overlay\" = accepted derived columns, corrections and mappings merged in, each row with `_row_key` (use it in propose_*) and `_overlay` (original value, reason, who, when); \"proposed\" (owner only) = also the open proposals, each with its state. Not combinable with aggregate or summary_only.",
+  "enum": [
+    "source",
+    "with_overlay",
+    "proposed"
+  ],
+  "type": "string"
+}
  2. Changed2 schema fields changed
    • changedInput schema / properties / format / description
      Previous value: -"Output wire format. 'compact' (default) = minified JSON, token-efficient. 'json' = pretty JSON for readability. 'toon' = Token-Oriented Notation, fewest tokens for tabular rows."New value: +"Output wire format for this MCP call. Default 'toon' (Token-Oriented Notation, fewest tokens, best for tabular rows). 'compact' = minified JSON. 'json' = pretty JSON for readability. The REST API always returns JSON regardless."
    • changedInput schema / properties / format / enum
      Previous value: -[
-  "compact",
-  "json",
-  "toon"
-]New value: +[
+  "toon",
+  "compact",
+  "json"
+]
  3. Changed1 schema field changed
    • addedInput schema / properties / format
      Added value: +{
+  "description": "Output wire format. 'compact' (default) = minified JSON, token-efficient. 'json' = pretty JSON for readability. 'toon' = Token-Oriented Notation, fewest tokens for tabular rows.",
+  "enum": [
+    "compact",
+    "json",
+    "toon"
+  ],
+  "type": "string"
+}
  4. Changed3 schema fields changed
    • changedInput schema / properties / aggregate / description
      Previous value: -"Comma-separated aggregations as \"func(column)\". Functions: avg, sum, count, min, max, stddev, median. Example: \"avg(value),count(*),max(price)\". Result columns are aliased as func_col (e.g. avg_value)."New value: +"Comma-separated aggregations as \"func(column)\". Functions: avg, sum, count, min, max, stddev, median. Example: \"avg(value),count(*),max(price)\". Result columns are aliased as func_col (e.g. avg_value). Numerically correct on text-stored values."
    • addedInput schema / properties / non_null_only
      Added value: +{
+  "default": false,
+  "description": "Drop rows whose value is null or storage junk (datasets are often mostly empty). Use to avoid wasting tokens on null rows. Default false.",
+  "type": "boolean"
+}
    • addedInput schema / properties / summary_only
      Added value: +{
+  "default": false,
+  "description": "Return only a finished per-column stats block (n, min, max, avg) plus first/last endpoint values, and NO raw rows. Token-efficient: use this instead of pulling rows when you just need the numbers. Default false.",
+  "type": "boolean"
+}
  5. First observed

TDQS

A4.4/5.0
Behavior4/5

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

Annotations already declare the safe read-only/idempotent profile, and the description adds non-obvious behavior the annotations cannot: aggregates are numerically correct on text-stored values (no lexicographic min/max), and the server refuses to combine columns of different kinds. Token-cost framing (17,000 raw rows vs ~200 tokens) is useful operational context. It loses a point for burying a hard constraint ('autario refuses to combine columns of different kinds') mid-paragraph and for a citation mandate unrelated to correct invocation.

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

Conciseness3/5

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

The core purpose and the get_entity_data split are front-loaded, which is good. But the 'semantics' instruction is stated twice in adjacent sentences ("Read the semantics field of the schema before combining two columns" and "Call get_dataset_schema first and read its `semantics` block"), summary_only is explained in the body and again in the schema, and the citation reminder is a non-sequitur that dilutes the call-critical content.

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

Completeness4/5

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

With 12 parameters, no output schema, and a genuine safety-relevant constraint (don't mix column kinds), the description covers the essentials: what comes back ("rows as JSON plus per-category statistics, or just the summary when summary_only"), how to minimize tokens, and which sibling to use instead. Missing is any guidance on pagination limits or what happens when filter/aggregate combinations are invalid, though the schema carries some of that.

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

Parameters4/5

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

Schema description coverage is already 100%, so the baseline is 3; the description earns above that by explaining the tradeoff behind summary_only and non_null_only rather than restating them, and by supplying the question-to-aggregate translation pattern. It does not, however, touch filter/sort/groupby syntax, leaving those entirely to the schema.

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?

Opens with a specific verb+resource ("Query the rows of ONE dataset") and immediately bounds scope by enumerating the source families (World Bank, FRED, GA4, Shopify, shared_with_you uploads). It explicitly distinguishes itself from the closest sibling, get_entity_data, so an agent can route without opening either 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?

Gives concrete when-to-use routing ("Reach for get_entity_data instead when you want ONE entity across SEVERAL indicators joined on time") and a prerequisite ("Call get_dataset_schema first"). It also supplies an explicit preference rule — prefer aggregate/summary_only over raw rows — with a worked mapping from a natural-language question to aggregate=avg(value) + filters.

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

All Autario 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