US Treasury yield curve
Top drivers
⌁ mcp.call("adw-054") vADW-054-live-1.0 How deep is the current yield-curve inversion and is it deepening?
US Treasury yield curve
Top drivers
⌁ mcp.call("adw-054") vADW-054-live-1.0 A credit-allocation agent ingests ADW-054 weekly; when inversion_signal_score exceeds 60 (current: 62.2, 78th percentile, backtest range 19.9–73.8) with a falling trend — meaning inversion is deep but starting to normalize, historically the window most correlated with recession onset — the agent reduces allocation to cyclical corporate credit and increases cash equivalents, logging methodology_version v0.1 and source_lineage (FRED T10Y2Y, T10Y3M) for the audit trail. The IOM's composite_z field lets the agent distinguish a true inversion from a mere rate-level move.
A bank's risk desk uses ADW-054 as a leading indicator for loan-loss provisioning reviews. At 62.2 (78th percentile of a 10-year history), both the 2-year and 3-month spreads are simultaneously inverted at above-average depth — a combination the NY Fed's free yield-curve model captures only at a one-month lag. The desk can front-run the quarterly CECL reserve cycle by flagging elevated recession probability now, rather than reacting after credit deterioration appears in charge-off data.
Negate both spreads → z_10y2y (0.5) + z_10y3m (0.5) vs trailing 130-obs window → composite z → 0-100 (50=neutral, >50=more inverted than hist avg)
Version ADW-054-live-1.0 · validated to beat a naive baseline · benchmark: NY Fed yield-curve recession model (free, lagged)
One call returns the answer with its reasoning attached — the live Intelligence Object for ADW-054.
{
"product_id": "ADW-054",
"entity": "US Treasury yield curve",
"score": 62.2,
"trend": "inversion-deepening",
"confidence": 0.88,
"top_drivers": [
{
"factor": "t10y2y_inverted_z",
"contribution": 1.083
},
{
"factor": "t10y3m_inverted_z",
"contribution": 0.137
}
],
"recommended_use": "Track yield-curve inversion depth as a recession-risk signal. High score = deep inversion historically associated with future downturns. Descriptive; no predictive IC claimed.",
"methodology_version": "ADW-054-live-1.0",
"freshness": "2026-06-26T07:00:10.784Z",
"coverage": "US Treasury yield curve (FRED T10Y2Y + T10Y3M)",
"source_lineage": [
"FRED T10Y2Y (10Y-2Y Treasury Spread)",
"FRED T10Y3M (10Y-3M Treasury Spread)"
],
"allowed_use": "informational",
"inversion_signal_score": 62.2,
"inversion_depth_label": "flat-or-normal",
"t10y2y_spread_pct": 0.31,
"t10y3m_spread_pct": 0.56,
"t10y2y_inverted_z": 2.165,
"t10y3m_inverted_z": 0.274,
"composite_z": 1.22,
"validation_status": "descriptive"
} Every product conforms to the Intelligence Object Model — typed, versioned, and discoverable.
Dashboard
Read the score + drivers in the console.
REST API
/v1/intelligence/adw-054
MCP tool
adw.adw_054
Marketplace
Discoverable by any MCP agent via the MCP registry.
White-label
Embed under your own brand (Platinum).
How healthy is the US macro economy right now?
Method: Z-score CPI/Unemployment/GDP; 1-(Inflation+Labor-Growth)
What is the stability & coverage of US banking?
Method: (Branch Density·Deposit Growth)/(Failure Rate·1000)
Are corporate credit spreads signaling elevated default risk right now?
Method: z_baa (0.5) + z_hy_oas (0.5); both z-scored vs trailing 130-observation window → composite z → 0-100 (50=neutral, >50=spreads wider than hist avg)