> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pylar.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Invoice & Billing Assistant

> Build an AI agent that handles billing queries, generates invoices, tracks payments, and manages collections

## Overview

An Invoice & Billing Assistant powered by Pylar handles billing inquiries, generates invoices, tracks payment status, and manages collections to streamline billing operations.

## What the Agent Needs to Accomplish

The agent must:

* Answer billing queries
* Generate invoices
* Track payment status
* Manage collections
* Identify overdue accounts
* Generate billing reports

## How Pylar Helps

Pylar enables the agent by:

* **Unified Billing View**: Combining invoices, payments, and customer data
* **Real-time Status**: Querying current billing and payment status
* **Automated Tracking**: Automated payment tracking
* **Collection Management**: Identifying and managing overdue accounts

```mermaid theme={null}
graph LR
    subgraph "Data Sources"
        INVOICES[Invoices]
        PAYMENTS[Payments]
        CUSTOMERS[Customer Data]
    end
    
    PYLAR[Pylar<br/>Unified Views]
    
    TOOLS[MCP Tools<br/>get_invoice<br/>track_payment<br/>manage_collections]
    
    AGENT[Billing Assistant]
    USERS[Users]
    
    INVOICES --> PYLAR
    PAYMENTS --> PYLAR
    CUSTOMERS --> PYLAR
    PYLAR --> TOOLS
    TOOLS --> AGENT
    USERS --> AGENT
    
    style PYLAR fill:#FF4017,stroke:#CC3300,color:#fff
```

## Without Pylar vs With Pylar

### Without Pylar

**Challenges**:

* ❌ Multiple billing systems
* ❌ Manual invoice generation
* ❌ Time-consuming payment tracking
* ❌ Limited collection visibility

**Implementation Complexity**: \~4-5 weeks

### With Pylar

**Benefits**:

* ✅ Single endpoint for billing data
* ✅ Automated invoice generation
* ✅ Real-time payment tracking
* ✅ Efficient collection management

**Implementation Complexity**: \~5-6 hours

## Step-by-Step Implementation

### Step 1: Connect Data Sources

1. **Connect Billing System** (Invoices, subscriptions)
2. **Connect Payment Processor** (Payments, transactions)
3. **Connect CRM** (Customer data, billing contacts)

### Step 2: Create Billing Views

**Invoice Status View**:

```sql theme={null}
CREATE VIEW invoice_status AS
SELECT 
  i.invoice_id,
  i.customer_id,
  i.invoice_date,
  i.due_date,
  i.amount,
  i.status,
  p.payment_amount,
  p.payment_date,
  i.amount - COALESCE(p.payment_amount, 0) as outstanding_amount,
  DATEDIFF(CURRENT_DATE, i.due_date) as days_overdue
FROM billing.invoices i
LEFT JOIN billing.payments p ON i.invoice_id = p.invoice_id;
```

### Step 3: Create MCP Tools

**Tool 1: Get Invoice Details**

* `get_invoice_details(invoice_id: string, customer_id: string)`

**Tool 2: Track Payment Status**

* `track_payment_status(invoice_id: string)`

**Tool 3: Identify Overdue Accounts**

* `identify_overdue_accounts(days_overdue: number, min_amount: number)`

**Tool 4: Generate Invoice**

* `generate_invoice(customer_id: string, items: array, due_date: date)`

## Example Agent Interactions

**User**: "What's the status of invoice #12345?"

**Agent**: "Invoice #12345 Status:

* Customer: Acme Corp
* Amount: \$5,000
* Status: Partially Paid (\$2,000 paid)
* Outstanding: \$3,000
* Due Date: March 15, 2024 (5 days overdue)
* Action: Send payment reminder"

## Outcomes

* **Query Response**: 80% faster
* **Payment Tracking**: 90% automation
* **Collection Efficiency**: 40% improvement
* **Customer Satisfaction**: 30% improvement

## Next Steps

* [Financial Analyst Example](/examples/financial-analyst)
* [Expense Auditor Example](/examples/expense-auditor)