Profile Dataset

Overview

When you send JSON rows, CSV text, or a file reference, we profile each column: inferred type, null counts, distinct counts, numeric min/max/mean/median/stddev, and IQR-based outlier candidates. We never execute CSV formulas or claim confirmed schema types. Do not use this for heavy statistical modeling or legal data-classification decisions.

Before you transform or migrate data, you need a fast, honest column summary — without guessing types in an LLM.

When to use it

Scope

Non-goals

• Does not execute CSV formulas
• Does not infer sensitive attributes or PII labels
• Type inference is heuristic — not a confirmed schema
• Outliers are IQR-based candidates, not confirmed anomalies
• Stats describe the profiled sample (up to sampleLimitRows), not necessarily the full file

Properties

Capabilities

Cost & billing

Flat rate per invocation

File handling

File limits

Constraints this skill applies to your input and generated files. The “Limited by” column shows the static source of each cap; none of these can be raised by upgrading your plan.

How we handle your data

Input schema

Output schema

API contract

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"data.profile_dataset","input":{"rows":[{"age":30},{"age":40}]}}'

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("data.profile_dataset", {"rows":[{"age":30},{"age":40}]});

Errors

Other limits

Limitations

• Does not execute CSV formulas
• Does not infer sensitive attributes or PII labels
• Type inference is heuristic — not a confirmed schema
• Outliers are IQR-based candidates, not confirmed anomalies
• Stats describe the profiled sample (up to sampleLimitRows), not necessarily the full file
• Heavy statistical modeling
• Legal or compliance classification decisions

How to call

POST /v1/execute (requires authentication)

{
  "skillKey": "data.profile_dataset",
  "input": {
    "rows": [
      {
        "age": 30
      },
      {
        "age": 40
      }
    ]
  }
}

Examples

{
  "rows": [
    {
      "age": 30
    },
    {
      "age": 40
    }
  ]
}

{
  "rowCount": 2,
  "columns": [
    {
      "name": "age",
      "inferredType": "integer",
      "nullCount": 0,
      "nullRatio": 0,
      "distinctCount": 2,
      "min": 30,
      "max": 40,
      "mean": 35,
      "median": 35,
      "stddev": 5,
      "outliers": [],
      "examples": [
        30,
        40
      ]
    }
  ],
  "profiledRowCount": 2,
  "columnCount": 1
}

{
  "includeExamples": false,
  "rows": [
    {
      "age": 30
    },
    {
      "age": 40
    }
  ]
}

{
  "rowCount": 2,
  "columns": [
    {
      "name": "age",
      "inferredType": "integer",
      "nullCount": 0,
      "nullRatio": 0,
      "distinctCount": 2,
      "min": 30,
      "max": 40,
      "mean": 35,
      "median": 35,
      "stddev": 5,
      "outliers": []
    }
  ],
  "profiledRowCount": 2,
  "columnCount": 1
}

Recommended use cases

• Understand a dataset before transformation
• Data quality spot-checks
• Compare profiledRowCount to rowCount before trusting stats on large files

Avoid for

• Heavy statistical modeling
• Legal or compliance classification decisions

Related skills

Composable with

Safety & privacy

• Runs without calling any external service.
• Your payloads are never stored permanently.
• Telemetry captures metadata only — never your payload contents.
• We never train models on your payloads.
• No human reviews your payloads.

Tags