📂

Analyze Dataset

Upload a CSV, JSON, Parquet, or Excel file URL and receive an AI-generated summary, schema, and preview rows.

POST /v1/analyze

curl -X POST "https://analytics.toolkitapi.io/v1/analyze" \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "data_url": "https://example.com/sales.csv",
    "prompt": "What are the top 5 products by revenue?",
    "file_type": "csv",
    "execution_mode": "sync"
  }'

import httpx

resp = httpx.post(
    "https://analytics.toolkitapi.io/v1/analyze",
    json={
    "data_url": "https://example.com/sales.csv",
    "prompt": "What are the top 5 products by revenue?",
    "file_type": "csv",
    "execution_mode": "sync"
  },
)
print(resp.json())

const resp = await fetch("https://analytics.toolkitapi.io/v1/analyze", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "data_url": "https://example.com/sales.csv",
    "prompt": "What are the top 5 products by revenue?",
    "file_type": "csv",
    "execution_mode": "sync"
  }),
});
const data = await resp.json();
console.log(data);

# See curl example

Response 200 OK

{
  "dataset_id": "ds_a1b2c3d4",
  "summary": "The dataset contains 1,200 rows of sales transactions. The top product by revenue is Widget A with $48,200.",
  "result_preview": [
    {"product": "Widget A", "total_revenue": 48200},
    {"product": "Widget B", "total_revenue": 31500}
  ],
  "schema_": [
    {"name": "product", "dtype": "string", "nullable": false},
    {"name": "total_revenue", "dtype": "float64", "nullable": false}
  ],
  "meta": {
    "request_id": "req_xyz789",
    "runtime_ms": 1240.5,
    "cache_hit": false,
    "rows_scanned_estimate": 1200,
    "schema_fingerprint": "fp_abc123"
  }
}

Try It Live

Description

Upload a CSV, JSON, Parquet, or Excel file URL and receive an AI-generated summary, schema, and preview rows.

How to Use

1

1. Upload your data file to a publicly accessible URL (or use an existing one). 2. Send a `POST` request to `/v1/analyze` with the URL and a plain-English `prompt`. 3. Use the `dataset_id` from the response in subsequent `/visualize` or `/save` calls. 4. For large files, set `execution_mode` to `async` and poll `/v1/jobs/{job_id}` for the result.

About This Tool

The **Analyze** endpoint is the entry point for all analytics workflows. Provide a publicly accessible URL pointing to a CSV, JSON, Parquet, or Excel file along with a natural-language prompt, and the API returns an AI-generated summary, a result preview, and the inferred schema.

The returned `dataset_id` is an in-memory handle to the analysed data. Pass it to `/visualize`, `/validate-chart`, or `/save` to continue the workflow without re-uploading or re-parsing the source file.

Why Use This Tool

Ad-hoc data exploration — ask a plain-English question and get an instant answer.
Schema discovery — call with a minimal prompt to infer column names and types before building a chart.
Pipeline automation — chain analyze → save → query to replay business logic against refreshed data.
BI prototyping — feed the result preview into a front-end chart builder without a full database.

Start using Analyze Dataset now

Get your free API key and make your first request in under a minute.

Get Free API Key View Docs