# Plaid Enterprise Banking MCP

> Plaid Enterprise Banking MCP connects your natural language AI directly to a comprehensive suite of financial intelligence tools. It lets you go beyond simple balance checks, giving your agent access to fraud risk scoring, detailed transaction history, investment holdings, and identity verification against global watchlists.

## Overview
- **Category:** money-moves
- **Price:** Free
- **Tags:** open-banking, financial-data, transaction-history, identity-verification, payment-processing, api-integration

## Description

This MCP gives your AI client the ability to function as a professional underwriter or forensic accountant by connecting it directly to Plaid's full API ecosystem. When you use this, your agent can read real-time account balances, analyze detailed transaction streams with merchant and geolocation data, and assess risk factors before transactions even happen using signals like ACH wire fraud detection. It also pulls critical information that goes beyond basic banking details; you can get a client’s current investment portfolio holdings or audit their outstanding student loan and credit card liabilities. For identity compliance, the MCP checks account holders against Interpol lists and OFAC sanctions. Connecting this capability through Vinkius means your agent gets instant access to best-in-class financial intelligence without needing specialized development work.

## Tools

### create_link_token
This tool generates the necessary secure token required to start connecting a user's bank account.

### exchange_public_token
It converts a temporary public token into the permanent access token needed for deeper data retrieval.

### get_accounts
Retrieves a list of all bank accounts linked and connected to the user's profile.

### get_balances
Fetches the most current, real-time financial balances across multiple linked accounts.

### get_categories
Lists and classifies all transaction types recorded in the user's history (e.g., groceries, utilities).

### get_identity
Pulls basic account holder identity details for verification purposes.

### get_institution
Retrieves specific identifying information about the bank or financial institution itself.

### get_item_info
Checks the connection status and details of an individual, linked data item within a connected account.

### get_transactions
Retrieves comprehensive transaction history, detailing every deposit and withdrawal made by the user.

### search_institutions
Searches a database to find financial institutions that support specific products or services.

## Prompt Examples

**Prompt:** 
```
Here is the access token for the client: `access-prod-1234`. Can you fetch their current credit card outstanding liabilities and highlight any accounts charging over 20% APR?
```

**Response:** 
```
Querying the Liability engine... The user has $4,200 total revolving debt. The 'Chase Sapphire' card holds $3,100 at 19.99% APR, which is fine. However, their 'Amex Platinum' currently carries $1,100 at a steep **24.5% APR**. We should advise a balance transfer.
```

**Prompt:** 
```
Investigate access token `access-prod-101` and check the investment brokerage holdings for AAPL and TSLA.
```

**Response:** 
```
Connecting to the brokerage via Plaid Wealth... I extracted 15 security holdings. The client holds 125 shares of Apple (AAPL) priced at $185/share, yielding $23,125 total value. For Tesla (TSLA), they hold a massive 400 share position yielding a valuation of $72,000. Do you want to see the capital gains on those?
```

**Prompt:** 
```
Using transaction access_token `access-prod-99`, analyze all ML recurring transaction signals. What subscriptions are they paying for?
```

**Response:** 
```
Running the Recurring Transactions Model... Plaid successfully detected 5 active passive flows on the main checking account: Netflix ($15.99), Spotify ($10.99), Equinox Gym ($205.00), AWS Cloud ($89.00), and an obscure app called 'AnyDo' ($4.99). Would you like to check the fraud signal prediction for these?
```

## Capabilities

### Review current balances and transaction streams
Your AI client can fetch real-time account balances, list connected accounts, and retrieve detailed transaction history with associated categories.

### Assess fraud risk on outgoing payments
The MCP evaluates the risk of ACH wires using Plaid's signal network before a payment processes, helping detect potential fraud.

### Audit investment and liability positions
It pulls live data on brokerage holdings, asset reports, and specific liabilities like credit card APRs or student loan balances.

### Verify identity against sanctions lists
Your agent can check the account holder's identity against major compliance watchlists like OFAC and Interpol.

### Process payroll data and bank routing numbers
The MCP parses raw W2 payroll stubs, verifies global employers, and safely extracts necessary 9-digit routing numbers for transfers.

## Use Cases

### Due Diligence on a Corporate Client
A risk manager needs to vet a new partner. They ask their agent to use the identity tools, check `get_accounts` for all connected entities, and then run fraud signal checks on any large outgoing transfers before approving the contract.

### Forensic Audit of Missing Funds
An accountant suspects unauthorized activity. They instruct their agent to retrieve the full `get_transactions` history, categorize every expense using `get_categories`, and then cross-reference it with the account holder's basic identity details.

### Financial Planning Review
A financial advisor runs a review by combining data from investment holdings (via Plaid Wealth) with liability reports. They can immediately identify if high APR student loan payments are draining capital needed for retirement savings.

### Onboarding and KYC Compliance
The onboarding team needs to verify a new client. They first use `create_link_token` and `exchange_public_token` to connect the account, then run identity verification using `get_identity` against global watchlists.

## Benefits

- You get a holistic client view. Instead of checking balances via `get_balances` and then pulling transaction details separately with `get_transactions`, your agent processes both in one context to spot trends like recurring payments.
- Compliance checks are instant. Checking identity compliance against Interpol or OFAC lists, which is normally a separate manual step, happens automatically when you use the identity tools.
- Go beyond simple banking. You can pull investment data and audit liabilities—things that require specific reports otherwise—just by calling the relevant functions on this MCP.
- Risk scoring is predictive. Before an ACH wire goes out, your agent evaluates fraud risk using Plaid's Signal AI network. This prevents losses before they hit your ledger.
- Data extraction is robust. You don’t just see numbers; you can run specialized tools to parse W2 payroll stubs and extract reliable routing numbers for transfers.

## How It Works

The bottom line is, you get instant, read-only access to massive volumes of financial and wealth data without ever needing to write a single line of integration code.

1. First, you connect your Plaid Client ID and Secret within the setup using this MCP.
2. Next, you pass a specific access token into your agent's prompts. The client uses that token to run targeted analyses against isolated account ledgers.
3. Finally, your AI client analyzes the returned data—be it transaction lists or liability reports—and presents actionable findings.

## Frequently Asked Questions

**Does Plaid Enterprise Banking MCP support payments?**
No. This MCP is strictly read-only. While it can inspect massive volumes of wealth and ML data, you cannot programmatically execute ACH debits or wires through this MCP.

**How do I get the transaction history using Plaid Enterprise Banking MCP?**
Use the `get_transactions` tool. It retrieves comprehensive records of every deposit and withdrawal, allowing your agent to analyze spending patterns over time.

**Can this MCP help with identity verification?**
Yes. The dedicated identity tools allow you to check an account holder's details against global sanctions lists like OFAC, which is critical for compliance.

**What if I need to connect a new bank account?**
You start by using `create_link_token` and then pass that token through `exchange_public_token` to get the secure access credentials needed for data retrieval.

**Is this MCP useful for investment tracking?**
Absolutely. You can pull real-time brokerage investment holdings and asset reports, giving you a view of wealth beyond just checking account balances.

**Is this safe to run against production Plaid accounts? Will the AI transfer money?**
Yes, it is entirely safe. We deliberately designed the 20 tools purely for deep inspection (fetching assets, checking IDs, balancing ledgers). Endpoints capable of moving capital (like `/transfer/create`) were excluded. The worst the AI can do is give you a highly detailed dashboard of your expenses.

**How does the agent analyze different bank accounts when I talk to it?**
You hold the master Server keys (Client ID and Secret). When you ask the agent a question, you just paste or mention the unique `access_token` representing that customer's bank connection, and the AI automatically uses it under the hood.

**Can it cross check watchlists (AML) and biometrics?**
Yes. Tools like `get_watchlist_screening` and `get_identity_verification` hook directly into Plaid Identity, matching end-users to PEP (Politically Exposed Persons), OFAC, and global government watchlists instantaneously.