Automated intelligence tool that crawls 32+ major bank websites, extracts tokenization initiatives using AI, and delivers structured data you can use.
In simple terms: it automatically finds and extracts bank tokenization news so you don't have to manually check 32 websites.
Automatically navigates bank websites, following links to find tokenization-related pages. Scores each page 0-100 for relevance.
GPT-4 reads each page and extracts structured data: offering type, status, currencies, use cases, and more.
SQLite database remembers what's been processed. Only NEW content gets extracted, saving you API costs.
Automatically generates a beautiful, SEO-optimized website from your data. Deploy anywhere.
Ready to crawl out of the box. Add your own banks easily via YAML config.
The tool runs from your terminal. Here's exactly what you do and see.
Download the tool, install Python packages, add your OpenAI API key to a .env file.
Execute a single command. Watch as it crawls bank websites and scores pages for relevance.
Pages and extracted offerings are saved to a local database file. Query it anytime.
One command creates a beautiful website, RSS feed, and all assets. Deploy to Netlify, Vercel, or any host.
Multiple formats to suit your needs: website, database, JSON, RSS feed.
A beautiful, responsive website showing all banks and their tokenization initiatives. Each bank gets a card with relevance score, status, and links to source articles.
Live Example: tokenews.netlify.app
Features: Bank cards, relevance badges, article links, RSS subscription, mobile-responsive, SEO-optimized.
Query with any SQLite tool or from Python:
-- Get all high-relevance pages SELECT bank_name, title, url, relevance_score FROM pages WHERE relevance_score >= 80 ORDER BY relevance_score DESC; -- Get pages for specific bank SELECT * FROM pages WHERE bank_id = 'jpm'; -- Get extracted offerings SELECT * FROM offerings WHERE bank_id = 'citi'; -- Count by region SELECT region, COUNT(*) as pages FROM pages GROUP BY region;
AI extracts structured data for each tokenization offering:
{
"bank_name": "J.P. Morgan",
"bank_id": "jpm",
"product_family": "Kinexys",
"offering_name": "JPM Coin",
"offering_type": "deposit_token",
"status": "Live",
"jurisdictions": ["US", "UK", "Singapore", "Germany"],
"currencies": ["USD", "EUR", "GBP", "SGD"],
"client_segments": ["Institutional", "Corporate"],
"blockchain_type": "Private/Permissioned",
"interoperability": {
"with_other_banks": ["DBS"],
"with_cbdc": "Exploring",
"with_stablecoins": "Supported"
},
"corporate_use_cases": [
"Cross-border payments",
"Intraday liquidity optimization",
"Programmable payments",
"Supply chain finance"
],
"sources": [
{"url": "https://www.jpmorgan.com/kinexys", "type": "official"}
]
}
Subscribe in any RSS reader to get updates:
<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
<channel>
<title>Bank Tokenization Monitor</title>
<link>https://tokenews.netlify.app/</link>
<description>Updates on bank tokenization</description>
<item>
<title>[100%] J.P. Morgan: Kinexys (JPM Coin)</title>
<link>https://www.jpmorgan.com/kinexys</link>
<description>Live - JPM Coin enables 24/7...</description>
</item>
<item>
<title>[100%] Citi: Citi Token Services</title>
<link>https://www.citigroup.com/...</link>
<description>Live - Instant settlement...</description>
</item>
</channel>
</rss>
Run the tool manually or set it to run automatically on a schedule.
Add to your server's crontab to run daily, weekly, or any schedule:
Included workflow file runs automatically. Just add your API key to GitHub Secrets.
Deploy to AWS Lambda, Google Cloud Functions, or Azure Functions with scheduled triggers.
The tool is terminal-based, but a UI can be added if needed.
| Feature | Terminal (Included) | UI (Optional Add-on) |
|---|---|---|
| Run crawls | ✓ Command line | ✓ Click a button |
| View results | ✓ Database queries, generated website | ✓ Dashboard with charts |
| Configure banks | ✓ Edit YAML file | ✓ Form interface |
| Schedule runs | ✓ Cron / GitHub Actions | ✓ Built-in scheduler |
| Technical skill needed | Basic command line | None |
| Cost | Included | +$199 (custom development) |
Most users are comfortable with the terminal. If you need a point-and-click UI, we can build one using Streamlit for an additional fee.
This tool requires an OpenAI API key that YOU obtain and pay for. We do not provide API keys or cover any API costs. Depending on how often you run the tool, expect to pay $50-500/month directly to OpenAI. Start with a single bank to estimate your costs before running full cycles.
| Usage Pattern | Estimated Monthly API Cost |
|---|---|
| 1 bank, once per week | $5 - $15 |
| All 32 banks, once per week | $30 - $80 |
| All 32 banks, daily | $50 - $150 |
| All 32 banks, every 4 hours | $150 - $450 |
One price. Everything included. No subscriptions.
Optional: Setup assistance for $199
pip install and python3 -m src.main. No coding required for normal use. If you want a point-and-click UI instead, we can build one for $199 extra.
SELECT url FROM pages WHERE relevance_score > 80, (3) Subscribe to the RSS feed in your favorite reader.
config/banks.yml file. Add the bank name, website domain, and starting URLs. The tool will crawl from there.
--bank flag to crawl specific banks only. Fewer banks = lower API costs.