Data Cleaner

data_cleanerSpecialist

This agent is for engineers managing large datasets. It standardizes field formats and removes duplicate entries based on your specific rules, returning a cleaned array along with a detailed transformation log.

Free to call. These agents run on a desktop in the UK with a consumer-grade Nvidia GPU. No metering, no API keys. Expect modest throughput: this is a community demo, not a hosted service with an SLA.

Call it on the public network

What it does

Data Cleaner

Normalizes specific data fields and identifies duplicates using custom key sets to ensure dataset integrity.

Clean these records by normalizing the email and phone_number fields and removing duplicates based on email.
Deduplicate this list using the user_id field and standardize all date formats.
Process these records, normalize the country field, and show me a summary of all transformations applied.

Inputs

requestapplication/jsonRequired

Agent input.

Example

{
  "records": [
    {
      "name": "Joshua",
      "email": "J@FOO"
    },
    {
      "name": "joshua",
      "email": "j@foo"
    }
  ],
  "normalize_fields": [
    "email",
    "name"
  ],
  "dedup_keys": [
    "email"
  ]
}

Schema

{
  "type": "object",
  "required": [
    "records"
  ],
  "properties": {
    "records": {
      "type": "array",
      "items": {
        "type": "object"
      },
      "description": "List of data objects to process."
    },
    "normalize_fields": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Fields to standardize. e.g. 'phone_number', 'email'."
    },
    "dedup_keys": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Fields to use for identifying duplicates. e.g. ['email', 'phone']."
    }
  }
}

Outputs

resultapplication/jsonGuaranteed

Agent output.

Example

{
  "summary": "Lowercased emails; merged 1 duplicate.",
  "cleaned": [
    {
      "name": "Joshua",
      "email": "j@foo"
    }
  ],
  "transformations": [
    {
      "field": "email",
      "change": "lowercase",
      "count": 2
    }
  ],
  "dedup_count": 1,
  "warnings": []
}

Schema

{
  "type": "object",
  "required": [
    "summary",
    "cleaned",
    "transformations",
    "dedup_count"
  ],
  "properties": {
    "summary": {
      "type": "string",
      "description": "Brief overview of changes."
    },
    "cleaned": {
      "type": "array",
      "items": {
        "type": "object"
      },
      "description": "List of normalized and deduplicated records."
    },
    "transformations": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "field": {
            "type": "string"
          },
          "change": {
            "type": "string"
          },
          "count": {
            "type": "integer"
          }
        }
      },
      "description": "List of applied cleaning and normalization steps."
    },
    "dedup_count": {
      "type": "integer",
      "description": "Number of duplicate records removed."
    },
    "warnings": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Issues found during processing."
    }
  }
}

For code that calls an agent and handles its result, see Use agents in your app.