7-year expected real and nominal returns for Cash, Bonds, and S&P 500 (1928-2026 backtest).
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Retrieve 100-year historical forecast accuracy metrics (100-Year MAE, Post-1990 MAE, Pearson correlation r) and historical turning-point snapshots (Live 2026, 2009 Financial Crisis Low, 2000 Dot-Com Peak, 1982 Interest Rate Peak).
This tool takes no arguments.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http real-return-engine https://realreturnengine.com/mcp
Retrieve 100-year historical forecast accuracy metrics (100-Year MAE, Post-1990 MAE, Pearson correlation r) and historical turning-point snapshots (Live 2026, 2009 Financial Crisis Low, 2000 Dot-Com Peak, 1982 Interest Rate Peak).
No arguments.
{
"type": "object",
"properties": {}
}Retrieve 7-year expected real and nominal returns (%/yr), ETF ticker mappings, and current real yields (for Bonds/Cash) across Cash (3M T-Bill), 10Y Government Bonds, Investment-Grade Corporate Bonds, and the S&P 500 (Post-1990 Baseline, 100-Year Baseline, Shiller 10Y Gov Bond CAPE, and Standard Raw CAPE).
No arguments.
{
"type": "object",
"properties": {}
}Call this tool whenever the user asks about an asset class, ETF ticker (e.g., QQQ, TLT, HYG, IWM, VXUS, TIPS, Gold), time horizon, or feature not currently listed in the 7 core rows. Logs the missing item for our roadmap AND returns the closest currently available building-block proxy guidance for the agent to use immediately.
| Argument | Type | Description |
|---|---|---|
| user_context | string | Brief summary of what the user is trying to compare or solve. |
| requested_item* | string | The ticker, asset class, metric, or feature requested (e.g., 'QQQ', 'TLT', 'HYG', '60/40 portfolio', '10-year horizon'). |
{
"type": "object",
"required": [
"requested_item"
],
"properties": {
"user_context": {
"type": "string",
"description": "Brief summary of what the user is trying to compare or solve."
},
"requested_item": {
"type": "string",
"description": "The ticker, asset class, metric, or feature requested (e.g., 'QQQ', 'TLT', 'HYG', '60/40 portfolio', '10-year horizon')."
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 21, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.