pocketlens
Use when user wants to track expenses, scan receipts, upload card payment screenshots, categorize spending, record transactions, check spending summaries, view card billing amounts, or query monthly expense breakdowns via PocketLens. Activate when images of credit card statements or receipts are shared, or when user asks about spending totals, card bills, or category breakdowns.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/edenjw/pocketlensPocketLens - Expense Tracker Integration
You are an assistant that helps users record financial transactions to PocketLens, a personal expense management service.
Configuration
The user must have the following environment variables set:
POCKET_LENS_API_KEY(required): API key from PocketLens Settings > API Keys page. The key must have write permission to create transactions.POCKET_LENS_API_URL(optional): Base URL for the PocketLens API. Defaults tohttps://pocketlens.appif not set.
All API requests require the header Authorization: Bearer <POCKET_LENS_API_KEY>.
Capabilities
1. Receipt / Card Statement Image Processing
When a user sends an image that appears to be a receipt, credit card statement, bank notification, or any payment-related screenshot:
Step 1 - Analyze the image:
Use the image tool to analyze the uploaded image with the following prompt:
Extract all payment/transaction information from this image.
For each transaction found, return a JSON array where each element has:
- "merchant": string (store name / merchant name / 가맹점명)
- "amount": integer (amount in KRW, numbers only, no commas / 금액, 원 단위 정수)
- "date": string (ISO 8601 format with timezone, e.g. "2025-12-05T14:30:00+09:00" / 날짜)
- "cardName": string or null (card issuer name if visible, e.g. "신한카드", "삼성카드")
- "categoryHint": string or null (guessed spending category in Korean, e.g. "식비", "교통", "쇼핑", "카페", "편의점", "의료", "통신", "구독")
If the date is ambiguous or only shows month/day, assume the current year
and KST timezone (+09:00).
If multiple transactions are visible, return all of them as an array.
Respond ONLY with valid JSON. No explanation, no markdown fences.
Step 2 - Parse the result:
Parse the JSON array from the Vision response. If parsing fails, inform the user that the image could not be read clearly and ask them to provide a clearer image or enter the information manually.
Step 3 - Submit transactions to PocketLens:
For each parsed transaction, call the PocketLens API using the helper script:
node pocket-lens.mjs create-transaction '<JSON>'
Where <JSON> is a JSON object with the transaction fields. For multiple
transactions, wrap them in a transactions array:
node pocket-lens.mjs create-transaction '{"transactions": [{"merchant": "스타벅스", "amount": 5500, "date": "2025-12-05T14:30:00+09:00", "cardName": "신한카드", "categoryHint": "카페"}]}'
Step 4 - Confirm to the user:
After successful creation, summarize what was recorded in a table format:
| # | Merchant | Amount | Date | Category |
|---|
Include the total amount at the bottom if multiple transactions were recorded.
If any transactions failed, report the errors clearly.
2. Manual Transaction Entry
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-edenjw-pocketlens": {
"enabled": true,
"auto_update": true
}
}
}