Methodology
How KabuWire Works
KabuWire operates a fully automated content pipeline that runs on a daily schedule. Here’s how each article is produced:
- Data Fetch — Real-time market data is retrieved from licensed financial data APIs (Finnhub, Alpha Vantage, FMP, FRED) at scheduled intervals.
- AI Generation — The data is injected into structured prompt templates and processed by Claude (Anthropic), an AI language model, to produce a readable article. All numbers in the article come directly from the fetched data — the AI formats and contextualizes, never invents.
- Validation — Generated articles are validated for length, required fields, and content quality before publication.
- Publication — Approved articles are published to WordPress and submitted to Google for indexing.
Editorial Standards
- All financial figures are sourced directly from market data APIs — never from AI training data
- Every article includes a disclaimer that it is not investment advice
- No buy/sell recommendations or price targets are ever published
- Articles report what happened — they do not predict future market direction
Data Sources
| Provider | Data Type |
|---|---|
| Finnhub | Stock quotes, earnings calendar, market news |
| Alpha Vantage | Market movers, sector performance |
| Financial Modeling Prep | Stock screener, dividend data, company profiles |
| FRED (Federal Reserve) | CPI, unemployment, GDP, Fed funds rate |