Expense Categorizer

expense_categorizerSpecialist

This agent is for small business owners needing to organize bookkeeping data. It processes raw transaction lists and maps them to specific accounts or a provided chart of accounts. It returns a categorized list, a summary, and flags any unidentifiable expenses.

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

Expense Categorizer

Matches transactions to account categories using vendor recognition and keyword analysis while assigning confidence scores.

Categorize these raw transactions using my provided chart of accounts.
Sort this list of expenses and flag anything you aren't sure about.
Map these bank statements to my business categories and give me a summary.

Inputs

requestapplication/jsonRequired

Agent input.

Example

{
  "transactions": [
    {
      "description": "Adobe Cloud",
      "amount": 52.99
    }
  ],
  "chart_of_accounts": [
    "Software",
    "Office",
    "Travel",
    "Meals"
  ]
}

Schema

{
  "type": "object",
  "required": [
    "transactions"
  ],
  "properties": {
    "transactions": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "description",
          "amount"
        ],
        "properties": {
          "date": {
            "type": "string"
          },
          "description": {
            "type": "string"
          },
          "amount": {
            "type": "number"
          },
          "vendor": {
            "type": "string"
          }
        }
      },
      "description": "List of raw transactions to categorize."
    },
    "chart_of_accounts": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Optional list of allowed categories."
    }
  }
}

Outputs

resultapplication/jsonGuaranteed

Agent output.

Example

{
  "summary": "12 categorized, 1 unknown.",
  "categorized": [
    {
      "description": "Adobe Cloud",
      "amount": 52.99,
      "category": "Software",
      "confidence": "high"
    }
  ],
  "unknown": []
}

Schema

{
  "type": "object",
  "required": [
    "summary",
    "categorized"
  ],
  "properties": {
    "summary": {
      "type": "string",
      "description": "Brief overview of categorized transactions."
    },
    "categorized": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "description",
          "category",
          "confidence"
        ],
        "properties": {
          "description": {
            "type": "string"
          },
          "amount": {
            "type": "number"
          },
          "category": {
            "type": "string"
          },
          "confidence": {
            "type": "string",
            "enum": [
              "high",
              "medium",
              "low"
            ]
          },
          "note": {
            "type": "string"
          }
        }
      },
      "description": "List of categorized transactions."
    },
    "unknown": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "List of unassigned categories."
    }
  }
}

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