Market Eyes Live - stock ratings and mortgage-rate context

io.github.marketeyeslive由 GitHub 帳號 marketeyeslive 發布v1.4.0更新於 Oct 24, 2026

Free stock-rating MCP: tier, 0-100 score, 8 factors, 11,000+ US tickers, plus mortgage rates.

已驗證Streamable HTTP可網頁執行金融資料與分析

概覽

AI 產生的概覽

一個免費、唯讀的遠端 MCP 伺服器，讓助理取得 Market Eyes Live 對美股與 ETF 的信心評級，以及每日美國房貸利率背景。

可以做什麼
查詢單一美股或 ETF 的信心等級、0 到 100 綜合評分、八項因子分數、風險提示與主題背景，並附資料日期與方法說明連結。將 2 至 5 檔美股或 ETF 並排比較等級、綜合評分以及估值、品質、動能分數。讀取當日房貸利率背景：10 年期美債殖利率與方向、MBS 動能、利率環境以及影響房貸利率的因素。每筆結果都帶有來源與免責聲明。
適用情境
適合使用者想快速了解某檔美股或 ETF 的演算法評級、需要比較幾檔標的，或詢問目前房貸利率環境時使用。它只提供研究資料，不構成建議，也不執行交易。
執行需求
遠端 Streamable HTTP 端點，無需安裝、API 金鑰或身分驗證。可用於 Claude Code、claude.ai、Claude Desktop、ChatGPT、Gemini CLI、Cursor、VS Code 等用戶端，也可透過純 JSON-RPC 呼叫。需要能連線至該端點的網路。
安裝前請注意
唯讀且從不執行交易，但評級屬於演算法研究資料，不構成投資建議，且可能變動。涵蓋範圍僅限美股與 ETF；加密貨幣、期貨及非美國上市標的會回傳未涵蓋結果。工具呼叫依使用者、平台及全域設有速率限制，結果中包含指向服務商網站的連結。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Market Eyes Live - stock ratings and mortgage-rate context，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.marketeyeslive%2Fmarket-eyes-live/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "market-eyes-live-stock-ratings-and-mortg": {
      "type": "http",
      "url": "https://marketeyeslive.com/mcp"
    }
  }
}

README

Market Eyes Live MCP Server

A free, read-only Model Context Protocol (MCP) server that returns Market Eyes Live's MELANY ratings for U.S.-listed stocks and ETFs (more than 11,000 tickers on its coverage list), plus a daily read of U.S. mortgage-rate conditions.

https://marketeyeslive.com/mcp

Remote server, Streamable HTTP transport, stateless JSON responses. No API key, no install, free. Full documentation: marketeyeslive.com/mcp-server.

A rating is a conviction tier and a 0 to 100 composite score built from eight factor scores. Published ratings are recorded and graded daily against later market moves; the method and the record are public (methodology, raw feed).

Tools

ToolInputReturns
get_stock_rating{ "symbol": "NVDA" }Conviction tier (weakest to strongest: Unfavorable, Hold, Favorable, Highest Conviction; names without mature fundamentals carry Promising, Very Promising, Rising Star, Runner! or Catalyst Watch), 0 to 100 composite, the eight factor scores (valuation, quality, momentum, earnings, sentiment, catalyst, risk-adjusted, macro fit), flagged risks, theme context, as-of date
compare_stocks{ "symbols": ["NVDA", "AMD"] } (2 to 5)Side-by-side tiers, composites, and valuation, quality and momentum scores from the daily-refreshed set
get_mortgage_rate_context{}Today's public read: 10-year Treasury yield and direction, MBS momentum, the rate environment (falling, stable or rising), and what moves mortgage rates

All three tools are read-only (readOnlyHint: true), declare an outputSchema, and return conforming structuredContent alongside the text. Every result carries source and a disclaimer; stock results also carry links.rating_page and links.methodology.

Coverage is U.S.-listed stocks and ETFs only. Crypto, futures, and non-U.S. listings (for example SHOP.TO or BTC-USD) return a "not covered" result. Five U.S. tickers that share their symbol with a futures contract or a coin (CORN, GOLD, WTI, BTC, ETH) are not rated, and the result says so. Class shares work in either spelling (BRK-B or BRK.B). A ticker that is not on the coverage list returns a result saying so. The tier is always one of: Unfavorable, Hold, Favorable, Highest Conviction, Promising, Very Promising, Rising Star, Runner!, Catalyst Watch. ETFs and REITs use the same ladder, chosen by the tone of the fund's own label in the app, and a leveraged or inverse fund never reads above Hold.

Connect

Claude Code:

claude mcp add --transport http market-eyes-live https://marketeyeslive.com/mcp

claude.ai and the Claude desktop app: Settings, then Connectors, then add a custom connector with URL https://marketeyeslive.com/mcp and no authentication.

ChatGPT: Plugins, then +, then Add custom MCP server, URL https://marketeyeslive.com/mcp, authentication none (Plus, Pro, Business, Enterprise and Edu plans; some accounts need developer mode). Menu names may vary.

Gemini CLI (this repository is a Gemini CLI extension; gemini-extension.json points at the endpoint over Streamable HTTP):

gemini extensions install https://github.com/marketeyeslive/market-eyes-live-mcp

Perplexity, Gemini, Grok, Mistral Le Chat: see the setup steps.

Set it up on a computer; once added, it also works in the phone apps of Claude and ChatGPT.

Cursor (.cursor/mcp.json):

json
{ "mcpServers": { "market-eyes-live": { "url": "https://marketeyeslive.com/mcp" } } }

Claude Desktop config file and VS Code:

json
{ "mcpServers": { "market-eyes-live": { "type": "http", "url": "https://marketeyeslive.com/mcp" } } }

Plain HTTP (JSON-RPC 2.0):

bash
curl -s https://marketeyeslive.com/mcp \  -H 'Content-Type: application/json' \  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_stock_rating","arguments":{"symbol":"AAPL"}}}'

Example structuredContent (illustrative values)

json
{  "symbol": "AAPL",  "rating_tier": "Highest Conviction",  "composite_score": 72,  "scale": "0-100, higher is stronger",  "as_of": "2026-10-06",  "factors": { "valuation": 41, "quality": 88, "momentum": 72, "earnings": 66, "sentiment": 58, "catalyst": 52, "risk_adjusted": 63, "macro_fit": 60 },  "top_risks": ["Valuation stretched vs 5-year average"],  "source": "Market Eyes Live",  "disclaimer": "Algorithmic research data, not financial advice and not personalized to anyone. Market Eyes Live does not place trades.",  "links": {    "rating_page": "https://marketeyeslive.com/stock/AAPL?utm_source=mcp&utm_medium=agent&utm_campaign=AAPL",    "methodology": "https://marketeyeslive.com/how-melany-is-tested.html?utm_source=mcp&utm_medium=agent&utm_campaign=methodology"  }}

Limits

Only tool calls count; connecting, listing tools and pings are free.

  • 240 tool calls per hour per user when the AI platform passes an anonymous user id (ChatGPT does), otherwise per network address.
  • 3,000 per hour shared by a platform that passes no user id (for example Claude).
  • 3,000 per hour shared by all callers that do not come through a platform.
  • 6,000 per hour across all callers.
  • Live scoring of a ticker outside the daily-refreshed set: 20 per hour per user or address, 100 per hour per platform, 150 per hour for callers that do not come through a platform, 300 per hour across all callers. Live scores are reused for 12 hours.

A limit hit returns an error result that names the limit and when it resets. For higher volume, email [email protected]. Terms: marketeyeslive.com/api-terms.

Disclaimer

Market Eyes Live provides algorithmic research data and educational tools, not financial advice. Ratings can change and are not a recommendation to buy or sell any security. Markets involve risk, including loss of principal. The server is read-only and never places trades.

來源：README.md，提交 b07fafa

工具

3

get_stock_rating

Get stock rating

Returns the Market Eyes Live MELANY rating for one U.S.-listed stock or ETF: a conviction tier (weakest to strongest: Unfavorable, Hold, Favorable, Highest Conviction; names without mature fundamentals carry Promising, Very Promising, Rising Star, Runner! or Catalyst Watch instead; ETFs and REITs use the same ladder), a 0 to 100 composite score, eight factor scores (valuation, quality, momentum, earnings, sentiment, catalyst, risk-adjusted, macro fit), flagged risks, theme context and the as-of date. Use this when the user asks how Market Eyes Live rates a U.S. ticker or wants a factor breakdown of one, for example 'what is the Market Eyes Live rating on AAPL' or 'how does NVDA score on valuation and momentum'. Do not use for real-time or intraday quotes, price history, options, order books, crypto, futures, non-U.S. listings, or a personal recommendation; tickers outside U.S.-listed stocks and ETFs return a not-covered message. Limitations: the daily-refreshed set answers instantly; other covered tickers are scored live at a lower hourly limit; ratings are algorithmic, dated by as_of, and can lag news; it does not return entry zones, alert levels or position sizing (when Market Eyes Live has a trade plan for the ticker, the trade_plan field says so; the plan values are part of a Market Eyes Live subscription). Methodology and the graded public record: https://marketeyeslive.com/how-melany-is-tested.html. Research data, not financial advice; this tool is read-only and never places trades.

輸入參數symbol必填

compare_stocks

Compare stocks

Lines up 2 to 5 U.S.-listed stocks or ETFs side by side on Market Eyes Live MELANY ratings: conviction tier (weakest to strongest: Unfavorable, Hold, Favorable, Highest Conviction), 0 to 100 composite score, and the valuation, quality and momentum factor scores. Use this when the user wants several tickers' ratings shown together, for example 'compare AAPL, MSFT and GOOGL on quality and valuation' or 'NVDA vs AMD ratings'. Do not use for real-time quotes, price history, options, crypto, futures, non-U.S. listings, or a personal recommendation. Limitations: draws only on the daily-refreshed rated set; a ticker outside it returns a note instead of scores (get_stock_rating can score a covered ticker live), and a ticker outside U.S.-listed stocks and ETFs returns a not-covered note; it does not return entry zones, alert levels or position sizing (has_trade_plan marks a ticker for which Market Eyes Live has a trade plan; the plan values are part of a Market Eyes Live subscription). Methodology and the graded public record: https://marketeyeslive.com/how-melany-is-tested.html. Research data, not financial advice; this tool is read-only and never places trades.

輸入參數symbols必填

get_mortgage_rate_context

Get mortgage rate context

Returns today's public U.S. mortgage-rate context from Market Eyes Live: the 10-year Treasury yield and its direction, mortgage-backed securities (MBS) momentum, the overall rate environment (falling, stable or rising), and a short explanation of what moves mortgage rates. Use this when the user asks what is driving U.S. mortgage rates or which way the rate environment is moving, for example 'what is driving mortgage rates this week' or 'are mortgage rates rising or falling'. Do not use for a specific lender's quoted rate, APR or fee comparisons, loan qualification, a personal lock decision, or anything outside the United States. Limitations: a daily read built from public market data; it describes conditions and does not forecast any individual's rate. Public market data and general education, not financial advice; this tool is read-only.

無需輸入參數。

版本歷史

2
  1. v1.4.0最新Oct 7, 2026
  2. v1.3.0Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分