American Default Research

What is American Default Research?

American Default Research is a Model Context Protocol server that exposes 96 economic distress indicators, the American Distress Index (ADI) composite score, and county-level distress scores across all 3,144 U.S. counties to MCP-compatible AI agents. It is designed for AI agents that need to retrieve structured economic distress data.

How to use American Default Research?

Point any MCP-compatible client at the hosted streamable-HTTP endpoint at https://mcp.americandefault.org/mcp. For Claude Desktop, add the endpoint URL to claude_desktop_config.json. The server exposes five tools: get_indicator, get_county_scorecard, get_adi_composite, search_indicators, and get_cross_correlations. A local install is possible for self-hosting but requires data files not included in the repo.

Key features of American Default Research

• Exposes 5 tools for retrieving distress data, ADI, and correlations.
• Data covers all 3,144 U.S. counties with 96 economic indicators.
• Each response includes canonical citations in APA, MLA, Chicago, and news-copy formats.
• Schema versioning ensures backward compatibility (v1 tools stay live).
• Built-in rate limiting for the HTTP transport (per-minute burst and per-hour sustained).
• Supports both stdio (for local clients) and streamable-HTTP transports.

Use cases of American Default Research

• An AI agent retrieves the latest American Distress Index composite score using get_adi_composite.
• A financial analyst's assistant fetches a county-level distress scorecard for a given FIPS code.
• A researcher searches for economic indicators by keyword (e.g., "mortgage delinquency") via search_indicators.
• An agent analyzes leading/lagging relationships between two indicators via get_cross_correlations.
• A policy brief generator automatically cites each data point with pre-formatted APA citations.

FAQ from American Default Research

What data sources does American Default Research use?

Data is sourced from FRED, BLS, NY Fed Household Debt and Credit Report, ATTOM Data Solutions, Mortgage Bankers Association, American Bankruptcy Institute / Epiq Systems, and additional primary government and industry sources.

Is the server free to use for AI agents?

The hosted endpoint is available without an explicit cost mentioned; the server supports anonymous and issued bearer-token tiers. Data is free to use with attribution per the canonical attribution block.

Can I run American Default Research locally instead of using the hosted endpoint?

The local installation is possible but requires data files (data/ and site/src/data/) not included in the repo. You can mirror data from the public API or use the repo as a code reference only.

What happens if an indicator slug has no populated data yet?

10 of 96 bundles return status: "awaiting_population" with full metadata and a null latest_value—agents can discover the slug exists without receiving phantom data.

How are response sizes managed for LLM context budgets?

Each endpoint has a defined response size budget (e.g., get_indicator ≤ 16 KB, get_county_scorecard ≤ 25 KB). The raw 300+ point indicator series is omitted from get_indicator to keep context manageable; the full series is available at a separate API endpoint.