cycle-detector
Real estate cycle phase detection using price momentum, credit conditions, construction activity, and sentiment indicators. Classifies markets into Recovery, Boom, Bust, or Stabilization phases.
Install
mkdir -p .claude/skills/cycle-detector && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/15264" && unzip -o skill.zip -d .claude/skills/cycle-detector && rm skill.zipInstalls to .claude/skills/cycle-detector
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Real estate cycle phase detection using price momentum, credit conditions, construction activity, and sentiment indicators. Classifies markets into Recovery, Boom, Bust, or Stabilization phases.About this skill
Property Cycle Detector
Purpose
Detect the current phase of the real estate cycle for any country or city using a multi-indicator framework. The property cycle is one of the most predictable economic cycles (18-year average) — correct phase identification is the foundation of property investment, mortgage risk management, and construction sector analysis.
When to Use
- "Where are we in the property cycle?"
- "Is it time to buy or sell property?"
- "Property market overheating?"
- "Is the housing correction over?"
- "Which cycle phase is [City/Country] in?"
Phase Detection Framework
7-Indicator Model
| # | Indicator | Weight | Recovery | Boom | Bust | Stabilization |
|---|---|---|---|---|---|---|
| 1 | Real House Price Momentum (12M) | 25% | Flat to +3% | +5 to +15% | -5 to -15% | -3 to +3% |
| 2 | Price-to-Income vs LT Avg | 20% | Below avg | Above avg | Falling to avg | At avg |
| 3 | Housing Credit Growth (YoY) | 15% | Low, turning up | High, accelerating | Contracting | Turning positive |
| 4 | Building Approvals (YoY) | 15% | Low, stabilizing | High, rising | Collapsing | Bottoming |
| 5 | Months of Inventory | 10% | High, declining | Low (<3 months) | Rising rapidly | Peaking, then declining |
| 6 | Auction Clearance Rate | 10% | 50-60% (improving) | > 70% | < 50% | 50-60% (stabilizing) |
| 7 | Investor Activity (% of lending) | 5% | Low, turning up | High (>35%) | Very low | Low |
Phase Classification
Each indicator is scored -2 (bust signal) to +2 (boom signal). Weighted sum determines phase:
| Score Range | Phase | Signal |
|---|---|---|
| +8 to +14 | Boom (Late) | Reduce — speculative excess |
| +4 to +7 | Boom (Early-Mid) | Hold — momentum intact |
| -3 to +3 | Transition | Watch — phase change approaching |
| -7 to -4 | Bust (Early-Mid) | Avoid — correction underway |
| -14 to -8 | Bust (Late) / Stabilization | Screen for value — bottom forming |
Country Data Sources
| Country | House Price Index | Source |
|---|---|---|
| US | Case-Shiller National | FRED: CSUSHPISA |
| UK | UK HPI / Nationwide | ONS API |
| Australia | CoreLogic HVI | ABS / CoreLogic |
| New Zealand | REINZ HPI | REINZ / RBNZ |
| Canada | Teranet-National Bank HPI | CREA / Teranet |
| China | 70-City Index | NBS |
Process
# US example
python3 mcp-servers/fred-mcp/server.py get_series '{"series_id":"CSUSHPISA","limit":60}'
python3 mcp-servers/fred-mcp/server.py get_series '{"series_id":"HOUST","limit":60}'
python3 mcp-servers/fred-mcp/server.py get_series '{"series_id":"MORTGAGE30US","limit":24}'
# NZ example
python3 mcp-servers/stats-mcp/server.py get_series '{"series_id":"house-price-index"}'
Red Flags
- National averages mask city-level divergence — drill down to city for investment decisions
- The cycle can extend — Phase 2 (Boom) has lasted 15+ years in some markets (Australia 1991-2017 with brief interruptions)
- Policy intervention (macroprudential, tax) can accelerate or delay phase transitions
- The 18-year cycle is an empirical observation, not a law — treat as framework, not prediction