US Market Bubble Detector evaluates bubble risk with a revised Minsky and Kindleberger framework. It first requires measured data: CBOE put/call ratio, VIX, FINRA margin debt, S&P 500 breadth and IPO activity. Six indicators are scored mechanically from 0 to 2, and price acceleration is added from the ten-year return distribution.
A strict qualitative step adds at most three points and only with measurable evidence such as Google Trends multiples and dated media coverage. The final score maps to a phase and a risk budget. The report is an analytical framework and is not investment advice.
What You Get
- A report with a score out of 15 and a phase from Normal through Caution, Elevated Risk and Euphoria to Critical
- A table of six indicators with measured values, scores and rationale
- A qualitative section where each added point lists its evidence and passes a double-counting check
- A risk budget range and the checklist of conditions the framework describes for each phase
Common Use Cases
- Asking whether the market looks like a bubble and getting a scored answer with sources
- Reviewing valuation concerns against measured leverage, volatility and breadth data
- Comparing a past score with the current one using the same fixed criteria
What Makes It Unique
Qualitative points are capped at three and each needs measurable proof, with a checklist against double counting. The version history explains how an earlier, looser revision overscored on narrative alone. A bundled scorer and reference notes on past bubbles and a data collection guide back the method.
Under The Hood
The agent collects Phase 1 data with web search and by reading public pages such as CBOE, Yahoo Finance, FINRA, Barchart and Renaissance Capital, or uses values you supply. A bundled Python script, which uses only the standard library, scores the indicators mechanically. The agent writes the final Markdown report. The skill needs no API key, and it points to reference files for the framework, historical cases and a daily checklist.