account-lookup

name	account-lookup
description	Retrieve customer account details — balances, account type, status, and linked accounts using simulated banking data (Faker)
enabled	true

Instructions

When the user asks about their account balance, account details, account status, or linked accounts, use this skill to retrieve the information.

1. Data Retrieval

This skill uses the Faker MCP server (configured in .mcp.json) to generate realistic banking data. Call the Faker MCP tools directly to produce customer and account data — no need to write inline Python with Faker.

Faker MCP tools to use:

• faker_bothify — generate a Customer ID pattern (e.g., "CUST-####??" → CUST-4821AB)
• faker_numerify — generate masked account numbers (e.g., "****####" → ****4821)
• faker_date_between — generate account opened dates
• faker_city — generate branch names
• faker_random_element — pick account status, types, etc.
• faker_pyfloat — generate balances and interest rates

Example flow: Use the Faker MCP to generate individual fields, then assemble them into a structured account response. For complex assembly (combining multiple Faker outputs into a JSON structure), use code_interpreter with the Faker MCP outputs as inputs.

Expected output structure:

{
  "customer_id": "CUST-4821AB",
  "accounts": [
    {
      "account_number": "****4821",
      "account_type": "Checking",
      "status": "Active",
      "balance": 12450.00,
      "available_balance": 12250.00,
      "currency": "USD",
      "opened_date": "2021-01-15",
      "interest_rate": 0.0,
      "branch": "Springfield Branch"
    }
  ]
}

2. Response Format

Present account information in a clean, readable format:

Field	Checking (****4821)	Savings (****9173)
Status	Active	Active
Balance	$12,450.00	$43,200.75
Available	$12,250.00	$43,200.75
Interest Rate	—	3.25% APY
Opened	Jan 15, 2021	Mar 8, 2019

3. Security Rules

• Never show full account numbers — always mask as ****XXXX
• Never show SSN, full card numbers, or passwords
• Include a timestamp: "Balances as of [date/time]"

4. Proactive Suggestions

After showing account details, consider suggesting:

• If checking balance is high: "You might benefit from moving some funds to a high-yield savings account."
• If savings rate is low: "Our new High-Yield Savings offers up to 4.25% APY."
• If account is dormant: "Your account has been inactive. Would you like to learn about reactivation?"

Chaining

This skill works best when combined with:

• Faker MCP — generates realistic customer and account data
• code_interpreter — assembles Faker outputs into structured responses
• transaction-history — after viewing balances, users often want to see recent transactions
• product-catalog — suggest relevant products based on account state

Constraints

• All data is simulated using Faker — include a disclaimer on first use
• Always mask sensitive fields
• Include "Balances as of" timestamp