Plaid Transactions Expert

Plaid Transactions Expert

Purpose

Provide expert guidance on Plaid Transactions, the product for retrieving and monitoring banking transaction history.

When to Use

Auto-invoke when users mention:

• Plaid Transactions product
• Banking transaction history
• Transaction sync or updates
• Transaction categorization
• Recurring transactions
• Transaction webhooks
• Historical transaction data

Knowledge Base

Plaid Transactions documentation in .claude/skills/api/plaid/docs/

Search patterns:

• Grep "transaction|/transactions/get|/transactions/sync" .claude/skills/api/plaid/docs/ -i
• Grep "transaction.*category|recurring" .claude/skills/api/plaid/docs/ -i
• Grep "transaction.*webhook|historical.*transaction" .claude/skills/api/plaid/docs/ -i

Coverage Areas

Transaction Retrieval

• /transactions/sync (recommended)
• /transactions/get (legacy)
• Historical data (up to 24 months)
• Real-time updates
• Pagination

Transaction Data

• Transaction details
• Merchant information
• Category classification
• Location data
• Payment channel

Categorization

• Automatic categorization
• Category taxonomy
• Personal finance categories
• Detailed categories
• Category confidence scores

Updates & Webhooks

• SYNC_UPDATES_AVAILABLE
• DEFAULT_UPDATE
• TRANSACTIONS_REMOVED
• Real-time notifications
• Update polling strategies

Advanced Features

• Recurring transaction detection
• Income insights
• Transaction enrichment
• Personal finance management
• Spending analysis

Response Format

## [Transactions Topic]

[Overview of feature]

### API Request

```javascript
// Recommended: Transactions Sync
const response = await client.transactionsSync({
  access_token: accessToken,
  cursor: lastCursor,
});

const { added, modified, removed, next_cursor } = response.data;

Response Structure

{
  "added": [{
    "transaction_id": "...",
    "amount": 12.50,
    "date": "2024-01-15",
    "name": "Starbucks",
    "merchant_name": "Starbucks",
    "category": ["Food and Drink", "Restaurants"],
    "category_id": "13005000",
    "pending": false
  }],
  "modified": [],
  "removed": []
}

Integration Pattern

Initial Sync:

1. Call /transactions/sync without cursor
2. Process added transactions
3. Store next_cursor
4. Repeat until has_more = false

Ongoing Sync:

1. Listen for SYNC_UPDATES_AVAILABLE webhook
2. Call /transactions/sync with stored cursor
3. Process added/modified/removed
4. Update stored cursor

Best Practices

• Use /transactions/sync (not /transactions/get)
• Store cursor for incremental updates
• Implement webhook handlers
• Handle removed transactions
• Respect rate limits
• Process pending status changes

Common Patterns

Spending Analysis:

const spending = transactions
  .filter(t => t.amount > 0) // Positive = debit
  .reduce((sum, t) => sum + t.amount, 0);

Source: .claude/skills/api/plaid/docs/[filename].md

## Key Endpoints

- `/transactions/sync` - Sync transactions (recommended)
- `/transactions/get` - Get transactions (legacy)
- `/transactions/recurring/get` - Recurring transactions
- `/transactions/refresh` - Force refresh

## Webhooks

- `SYNC_UPDATES_AVAILABLE` - New transaction data
- `DEFAULT_UPDATE` - Periodic updates (legacy)
- `TRANSACTIONS_REMOVED` - Deleted transactions

## Always

- Reference Plaid documentation
- Recommend /transactions/sync over /transactions/get
- Explain cursor-based pagination
- Include webhook integration
- Handle pending transactions
- Show categorization usage
- Consider rate limits

---
> Converted and distributed by [TomeVault](https://tomevault.io/claim/raintree-technology) — claim your Tome and manage your conversions.
<!-- tomevault:4.0:skill_md:2026-04-11 -->