# BRAINIAC CSV Outlier Detection API

> BRAINIAC CSV Outlier Detection API is a paid API for AI agents from www.brainiac.software, paid per call via x402, $0.05/call, status unknown (last checked 2026-11-17).

Detects statistical outliers in tabular data (CSV, JSON, or records) using the Tukey 1.5×IQR fence method and returns per-column bounds and flagged rows

## Facts

- Endpoint: POST https://www.brainiac.software/api/v1/data-quality/csv-outlier-detect?utm_source=zero.xyz
- Price: $0.05/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-17
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/brainiac-csv-outlier-detection-api-68413486
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_8kEonz_dBrqtwFSrPzrdN

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability brainiac-csv-outlier-detection-api-68413486 -d '<json body>'
```

Example prompt: Can you check this CSV for statistical outliers? I want to know which rows have values that fall outside the normal range for each numeric column — here's the data: id,amount,status\n1,10.5,active\n2,500,active\n3,11.2,active

## When to prefer this

Use this endpoint when you need a fast, accountless, pay-per-call statistical outlier scan on tabular data without setting up a full data-quality pipeline. It is ideal for pre-import validation, one-off audits, or agent workflows that need programmatic anomaly flagging on CSV or JSON records. Prefer this over manual inspection or heavier analytics tools when the dataset fits within 2,000 rows and 40 columns and you want Tukey IQR-based detection with per-column bounds returned in a structured JSON response.

## Known failure modes

- Input exceeds 2,000 rows — rows beyond the limit are omitted and recorded in recordsOmitted
- Input exceeds 40 columns — extra columns are omitted and recorded in columnsOmitted
- Malformed CSV or JSON that cannot be parsed — returns an error status
- CSV or JSON string exceeds 200,000 character limit — rejected
- File base64 exceeds 12,000,000 character limit — rejected
- Columns with insufficient numeric data for IQR calculation are skipped (status: not_checked)
- Missing required payment header or insufficient USDC balance — payment fails with 402 response

## How this service works

Pay-per-call APIs for AI agents: OFAC sanctions screening, compliance checks, token security, and Base chain data via x402. No account. Fixed USDC price. MCP-ready.

## Output

A JSON object containing: per-column statistics (name, count of outliers, lower/upper Tukey fence bounds, check status), a list of flagged outlier rows with the offending value, column index, row number, the specific Tukey rule violated, and a human-readable detail string; plus summary fields like outlierCount, blankRows, recordsRead, columnsOmitted, recordsOmitted, and a checkedAt timestamp. Non-numeric and non-date columns are skipped and marked as not_checked.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "csv": {
   "type": "string",
   "maxLength": 200000
  },
  "file": {
   "type": "object",
   "required": [
    "mediaType",
    "fileBase64"
   ],
   "properties": {
    "filename": {
     "type": "string",
     "maxLength": 180,
     "minLength": 1
    },
    "mediaType": {
     "enum": [
      "text/csv",
      "application/json",
      "text/plain"
     ],
     "type": "string"
    },
    "fileBase64": {
     "type": "string",
     "maxLength": 12000000,
     "minLength": 1
    }
   },
   "additionalProperties": false
  },
  "json": {
   "type": "string",
   "maxLength": 200000
  },
  "records": {
   "type": "array",
   "items": {
    "type": "object",
    "propertyNames": {
     "type": "string",
     "maxLength": 80,
     "minLength": 1
    },
    "additionalProperties": {
     "anyOf": [
      {
       "type": "string",
       "maxLength": 4000
      },
      {
       "type": "number"
      },
      {
       "type": "boolean"
      },
      {
       "type": "null"
      }
     ]
    }
   },
   "maxItems": 5000
  },
  "keyColumns": {
   "type": "array",
   "items": {
    "type": "string",
    "maxLength": 80,
    "minLength": 1
   },
   "maxItems": 20
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "format": "csv",
  "issues": [],
  "status": "checked",
  "columns": [
   {
    "name": "id",
    "count": 12,
    "index": 1,
    "lower": "-7",
    "upper": "19",
    "status": "checked"
   },
   {
    "name": "email",
    "count": 0,
    "index": 2,
    "lower": "",
    "upper": "",
    "status": "not_checked"
   },
   {
    "name": "amount",
    "count": 12,
    "index": 3,
    "lower": "8.875",
    "upper": "13.875",
    "status": "checked"
   },
   {
    "name": "status",
    "count": 0,
    "index": 4,
    "lower": "",
    "upper": "",
    "status": "not_checked"
   },
   {
    "name": "joined",
    "count": 0,
    "index": 5,
    "lower": "",
    "upper": "",
    "status": "not_checked"
   }
  ],
  "outliers": [
   {
    "row": 9,
    "rule": "tukey_1_5_iqr",
    "value": "500",
    "column": "amount",
    "detail": "The value is outside the Tukey fence 8.875 to 13.875. This is a statistical outlier, not a confirmed error.",
    "confirmed": false,
    "sourceRow": 10,
    "columnIndex": 3
   }
  ],
  "blankRows": 0,
  "checkedAt": "2026-11-03T14:00:00.000Z",
  "limitations": [
   "No business system is contacted or updated.",
   "Rows past the 2,000-row limit and columns past the 40-column limit are not read. recordsOmitted and columnsOmitted state how many.",
   "Rows that were read stay in the counts. None of those rows is deleted.",
   "Normalization is a proposal. The original cell is kept beside the proposed value.",
   "A statistical outlier is not a confirmed error.",
   "A date is normalized only when it is already a real YYYY-MM-DD day."
  ],
  "recordsRead": 12,
  "dataModified": false,
  "outlierCount": 1,
  "columnsOmitted": 0,
  "recordsOmitted": 0,
  "recordsDiscarded": false,
  "outliersConfirmed": false
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/brainiac-csv-outlier-detection-api-68413486/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from www.brainiac.software](https://www.zero.xyz/host/www.brainiac.software/llms.txt)
