market-momentum-report

Date anchor: Today's date comes from the # currentDate system context. Compute ALL relative dates from it. Example: if today = 2026-03-14, then "prior month" = 2026-02-01 to 2026-02-28, "current month" (most recent complete) = February 2026, "three months ago" = December 2025. Never use training-data dates.

Market Momentum Report — Monthly Portfolio Risk & Opportunity Intelligence

Lender Profile (Load First)

Load the marketcheck-profile.md project memory file if exists. Extract: portfolio_focus, country, state, tracked_segments, risk_ltv_threshold, high_risk_ltv_threshold. If missing, produces national overview. US-only. Confirm profile.

User Context

Lender (auto finance director, residual value committee member, portfolio risk manager) needing a comprehensive sector-level view for lending policy, residual setting, and portfolio strategy. Broadest skill — covers entire US auto market through a lending risk lens.

Workflow: Monthly Market Momentum

Step 1 — Macro signals (total market)

Total volume: Call mcp__marketcheck__get_sold_summary with:

• date_from / date_to: current month
• ranking_dimensions: inventory_type
• ranking_measure: sold_count
• top_n: 5

Repeat for prior month and 3 months ago. → Extract only: sold_count, average_sale_price, average_days_on_market per inventory_type per period. Discard full response.

Calculate:

• Total units MoM %
• Avg transaction price MoM %
• New vs Used mix shift
• Industry-wide average DOM trend

EV penetration: Call with fuel_type_category=EV for current and prior. Calculate penetration rate and bps change. → Extract only: sold_count for EV per period. Discard full response.

Step 2 — Winners and losers (by market share)

Call mcp__marketcheck__get_sold_summary with:

• ranking_dimensions: make
• ranking_measure: sold_count
• ranking_order: desc
• top_n: 25
• Current month AND prior month → Extract only: make, sold_count per period. Discard full response.

Calculate share % and bps change for each make. Identify:

• Top 5 gainers (largest positive bps change) — brands with improving residual outlook
• Top 5 losers (largest negative bps change) — brands with deteriorating residual outlook

Step 3 — Pricing power index

Call mcp__marketcheck__get_sold_summary with:

• inventory_type: New
• ranking_dimensions: make
• ranking_measure: price_over_msrp_percentage
• ranking_order: desc
• top_n: 20
• Current month → Extract only: make, price_over_msrp_percentage per brand. Discard full response.

Categorize:

• Above MSRP (still commanding premiums): count and avg premium % — lower residual risk
• At MSRP (within +/-1%): count — standard residual assumptions
• Below MSRP (discounting): count and avg discount % — elevated residual risk on new originations

Track overall: what % of new vehicles sell above/below MSRP? Compare to prior month.

Step 4 — Residual risk alert (depreciation)

Call mcp__marketcheck__get_sold_summary with:

• inventory_type: Used
• ranking_dimensions: body_type
• ranking_measure: average_sale_price
• ranking_order: asc
• top_n: 10
• Current month AND 3 months ago → Extract only: body_type, average_sale_price per period. Discard full response.

Calculate monthly depreciation rate per segment. Flag segments with > 1.5%/month as "RESIDUAL RISK ACCELERATING -- tighten advance rates."

Also identify the 5 fastest depreciating specific models (by make/model):

• ranking_dimensions: make,model
• ranking_measure: average_sale_price
• ranking_order: asc
• top_n: 20 → Extract only: make, model, average_sale_price per period. Discard full response.

Cross-reference with 3-month-ago data.

Step 5 — Regional collateral variance (optional, if state data available)

Call mcp__marketcheck__get_sold_summary with:

• summary_by: state
• ranking_measure: average_sale_price
• ranking_order: desc
• top_n: 10 → Extract only: state, average_sale_price, sold_count per state. Discard full response.

Identify:

• Premium collateral markets (highest values)
• Discount collateral markets (lowest values)
• Collateral value spread (highest vs lowest)

Step 6 — Supply health

Call mcp__marketcheck__search_active_cars with:

• car_type: new, stats: price,dom, rows: 0

And separately with car_type=used. → Extract only: num_found, stats.dom.mean per car_type. Discard full response.

Calculate:

• Total active new inventory nationally (or by state)
• Total active used inventory
• Implied days supply for each
• MoM inventory build/draw trend

Output

Present: portfolio risk composite headline (FAVORABLE/STABLE/DETERIORATING/MIXED), macro signals table, brand performance gainers/losers, pricing power index, residual risk alert with fastest depreciating segments/models, and 3 lending policy signals (residual setting, advance rates, portfolio exposure).

Composite Health Signal Logic (Lending Context)

• FAVORABLE: Volume up > 2%, pricing stable or rising, days supply < 60, depreciation rates below segment averages — residual risk low, standard lending parameters appropriate
• STABLE: Volume +/-2%, pricing +/-1%, days supply 50-75 — maintain current lending parameters
• DETERIORATING: Volume down > 2%, pricing falling, days supply > 75, depreciation accelerating — tighten advance rates, reduce residual forecasts, increase monitoring frequency
• MIXED: Conflicting signals (e.g., volume up but days supply building and depreciation accelerating) — differentiate by segment

Important Notes

• US-only: Requires get_sold_summary for sold data.
• Use the most recent COMPLETE month. If today is March 5, "current month" = February.
• This is the broadest report — keep it scannable. Credit committees and auto finance directors need the signal in 30 seconds.
• If the user has tracked segments in their profile, highlight those segments prominently and call out any risk signals specific to their portfolio focus.
• Frame every metric through a lending lens: volume = origination demand, pricing = collateral value, depreciation = residual risk, supply = liquidation risk.
• This report works well as a regular monthly deliverable. Suggest: "Want me to run this at the start of each month for your credit committee?"