Context7
Feeds AI the latest docs for libraries and frameworks before it answers, so code it writes does not use outdated, removed functions.
Description from the maker (original English)
Read-only U.S. household-finance research: economic indicators, county scores, ADI and citations.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
https://mcp.americandefault.org/mcpRun this in your terminal
claude mcp add --transport http research https://mcp.americandefault.org/mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"research": {
"url": "https://mcp.americandefault.org/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"research": {
"type": "http",
"url": "https://mcp.americandefault.org/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
https://mcp.americandefault.org/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Tool names and descriptions exactly as the server reported them on Nov 6, 2026. The maker wrote them, not us; we show them as plain text.
get_indicatorFetch a compact snapshot of an American Default economic indicator by slug. Returns latest value, unit, frequency, direction, pre-computed aggregates (period averages, extremes, sustained runs), editorial prose (when available), and canonical APA / MLA / Chicago / news-copy citat
get_indicator_v2Fetch an explicitly versioned compact indicator snapshot. Returns the latest value and canonical citation plus `trend_suppressed` and `series_breaks` comparison-safety fields. Callers must pass `response_version='v2'`; unsupported versions fail closed. The legacy `get_indicator`
get_county_scorecardFetch a county's County Distress Index (CDI) scorecard by 5-digit FIPS code. Returns the 0-100 score (the share of the other U.S. counties with a lower composite, rounded down; field composite_score), score label, national + state rank (national_rank_ordinal is filled only for th
get_adi_compositeFetch the latest quarterly reading of the American Distress Index (ADI) composite. Returns the composite score (0-100), band (1-5) with its label, the composite's rank in its own published history (rank_in_history.reading: quote this beside the band label), the methodology readin
search_indicatorsSearch the 121-indicator registry by keyword. Returns ranked matches (up to `limit`, default 10, max 50) with slug, branded name, underlying name, category, and canonical URL. Scoring is substring+prefix over slug, branded_name, name, and category — e.g. query 'savings' returns b
get_cross_correlationsFetch statistically-validated leading/lagging relationships for an indicator. Source: the five-filter leading-indicator scanner (cross-correlation → first-differenced CCF → multi-crisis validation → Granger causality → out-of-sample validation). Returns two lists: `as_leader` (pa
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
The registry name was verified against the domain americandefault.org (the registry checks domain ownership)
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Responded in 330 ms (HTTP 200)
No sign-in needed
Read 6 tool descriptions and found no hidden instructions to the AI
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 3 of 3 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
The GitHub repo last changed –, and the latest version (1.1.1) reached the registry 5 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
In the same category: Context7, Codebase Memory, Exa
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
Feeds AI the latest docs for libraries and frameworks before it answers, so code it writes does not use outdated, removed functions.
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Read-only research tools for U.S. household financial distress: economic indicators, county scorecards, and the American Distress Index, with dates, original publishers, and citations. The public service requires no account or API key for normal use. It does not access private homeowner records or expose the warehouse, arbitrary SQL, or financial transactions.
Endpoint: https://mcp.americandefault.org/mcp
Transport: Streamable HTTP. Public documentation: https://americandefault.org/press/mcp/
In Claude, open Customize, Connectors, and add a custom connector using that URL. In ChatGPT, use the custom MCP connection in developer mode where your account supports it. Directory publication is a separate platform review; this README does not claim that either platform has approved the listing.
Anonymous access allows 10 calls per minute and 60 per hour per IP block. Higher-volume users can request an issued key through the public documentation. The static JSON endpoints are the suitable route for permitted bulk downloads. HTTP clients reading the website or its JSON endpoints should send a descriptive User-Agent identifying their application. Cloudflare Pages may reject a library's default identity with error 1010; a descriptive application name works without impersonating a browser. The hosted MCP transport is a separate endpoint.
The main domain advertises the connector at https://americandefault.org/.well-known/ai-catalog.json. Its entry points to https://mcp.americandefault.org/mcp/server-card, using the experimental MCP Server Card format. This format describes identity and connection details; tools remain discoverable through the MCP protocol.
For clients using Cloudflare's earlier discovery convention, both domains serve /.well-known/mcp/server-card.json. That compatibility card includes the exact tool descriptions and input schemas returned by the running SDK. The main-domain route proxies only this public metadata from the MCP host, without forwarding cookies or authorization. Cards support CORS, HEAD and conditional ETag requests. The registry metadata supplies the same name and version used at initialization. Discovery metadata does not grant access or guarantee agent recommendations.
| Tool | Input | Result |
|---|---|---|
search_indicators | query, optional limit (maximum 50) | Matching indicators and public page links |
get_indicator_v2 | slug, response_version: "v2" | Latest reading, sources, citations, freshness and comparison warnings |
get_indicator | slug | Legacy v1 reading and citations |
get_county_scorecard | County fips | Score, domains, findings, dates and citations |
get_adi_composite | None | Latest published quarter, domains and historical context |
get_cross_correlations | Indicator slug | Validated research relationships; empty results are normal |
Prefer v2 when comparing indicators. A series break or a suppressed trend means readings may not be comparable. Historical indicators are not presented as current. Full histories, where available, are at https://americandefault.org/api/indicators/{slug}.json.
This public repository is a source reference for the hosted connector. The production service also uses shared formatting, index metadata and public data files maintained separately. Cloning this repository alone does not provide a complete local data installation. Connect to the hosted endpoint above for normal use; no local process, API key or data download is needed.
For Cursor, add this to the client's MCP server configuration:
{
"mcpServers": { "american-default-research": { "url": "https://mcp.americandefault.org/mcp" } } }
For Cline, set the transport explicitly in cline_mcp_settings.json:
{
"mcpServers": { "american-default-research": { "type": "streamableHttp", "url": "https://mcp.americandefault.org/mcp" } } }
All six tools are read-only. Invalid slugs or county codes return errors rather than fabricated results. The hosted HTTP health endpoint is /up.
Our analysis is free to use with attribution to American Default Research. Source data stays under each publisher's terms; cite the source named here. These research tools do not need conversation history, account credentials, private documents, or personal financial details. Request bodies and bearer credentials must not be logged. The site's privacy policy is https://americandefault.org/privacy/.