A secure, EU-hosted utility that automatically converts PDF bank statements into structured CSV, Excel, QuickBooks, and OFX files.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install statementedge
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install statementedge using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
StatementEdge is a high-performance, privacy-first parser designed to convert PDF bank statements into cleanly structured digital data. Fully integrated into the ecosystem of Openclaw Skills, it helps developers and finance teams automate accounting preparation by processing uploads from over 100 banks across 22 countries. The skill handles complex layouts natively using a vision-native approach that surpasses traditional template-based OCR mechanisms.
All operations are securely processed in the EU (Frankfurt region) with strict data handling controls, including zero AI training on your uploaded PDFs and automatic deletion within an hour. This makes it an ideal choice for workflows requiring absolute compliance under GDPR guidelines.
To configure this tool as part of your Openclaw Skills integration, declare the required key in your local environment.
# Required API key from statementedge.com
export STATEMENTEDGE_API_KEY="your_api_key_here"
# Optional base URL override (used only for self-hosted or staging endpoints)
export STATEMENTEDGE_BASE_URL="https://www.statementedge.com"
curl -s -X POST \\
-H "Authorization: Bearer $STATEMENTEDGE_API_KEY" \\
-F "file=@/path/to/statement.pdf" \\
https://www.statementedge.com/v1/convert | jq
StatementEdge organizes extracted bank statement outputs into structured fields. The JSON response conforms to the following schema structure:
| Attribute Name | Data Type | Purpose |
|---|---|---|
id |
String | Unique UUID for the parsed job |
status |
String | Conversion job status (queued, processing, done) |
reconcileStatus |
String | Balance math verification (reconciled, reconciled_with_flags, needs_review) |
statement.bank_name |
String | Identified bank name (e.g., AIB) |
statement.currency |
String | ISO 3-character currency string |
statement.opening_balance |
Decimal | The starting ledger balance |
statement.closing_balance |
Decimal | The final statement ledger balance |
statement.transactions |
Array | List of extracted transactions, mapping date_iso, description, amount, balance, direction, and flags |
You can request structured formats directly without manual parsing middleware:
csv: Raw comma-separated valuesxlsx: Native multi-sheet Excel workbookqbo: Intuit QuickBooks Online formatofx: Standard Financial Exchange formattingxero-csv / sage-csv: Pre-mapped formats for direct manual accounting system uploads.qbo and .xlsx directly from the API endpoint, removing the need for local file converters.Loading
An advanced stock and cryptocurrency analysis engine featuring multi-dimensional scoring, viral trend detection, automated portfolio tracking, and real-time rumor scanning.

An advanced, multi-dimensional stock and cryptocurrency analysis tool with rumor detection, sentiment analysis, and portfolio management capabilities.

A high-performance AI agent skill designed for live system querying, API mapping, and comprehensive environment governance within the DashClaw platform.

Goofish Search is an automated browser skill designed to search Xianyu/Goofish for secondhand products, apply strict merchant filtering, sort by price, and output optimized top results.

Ultimate Domain Payload provides a comprehensive meta-domain structural payload that eliminates systemic boundaries across 12 key life domains, allowing seamless cross-domain workflow orchestration via Universal Task OS.

An automated, layout-driven tool designed to generate self-contained, beautifully formatted HTML slide decks and editable PPTX presentations directly from natural language prompts.








































