Skip to content

Market Eyes Live MCP

Free keyless MCP server that returns Market Eyes Live's MELANY ratings for U.S.-listed stocks and ETFs - more than 11,000 tickers - plus a daily read of U.S. mortgage-rate conditions. Three read-only tools, structured output, and a public record: published ratings are graded daily against later market moves, and both the methodology and the validation feed are open.

Server type: Remote (Streamable HTTP at https://marketeyeslive.com/mcp)
Auth: None (free, keyless)
Docs: https://marketeyeslive.com/mcp-server
Tools: get_stock_rating, compare_stocks, get_mortgage_rate_context
Coverage: U.S.-listed stocks and ETFs only; crypto, futures and non-U.S. listings return a not-covered result
Category: Finance
Built by: Market Eyes Live

Why This Matters for Operators

Most market data tells you a number; this server tells you a conviction and shows its work. A rating is a tier plus a 0 to 100 composite built from eight factor scores (valuation, quality, momentum, earnings, sentiment, catalyst, risk-adjusted, macro fit), and every result carries a source, an as-of date and a disclaimer. For operators, the mortgage-rate tool is the daily-use piece: 10-year Treasury yield and direction, MBS momentum and the rate environment, useful for anything from real-estate decisions to pricing conversations. And because published ratings are graded against what actually happened, you can judge the signal instead of trusting it.

Tools & Capabilities

Tool Returns
get_stock_rating conviction tier (Unfavorable, Hold, Favorable, Highest Conviction, plus Promising, Very Promising, Rising Star, Runner! and Catalyst Watch for earlier-stage names), 0 to 100 composite, the eight factor scores, flagged risks, theme context and as-of date
compare_stocks side-by-side tiers, composites and valuation, quality and momentum scores for 2 to 5 symbols
get_mortgage_rate_context today's read: 10-year Treasury yield and direction, MBS momentum, rate environment (falling, stable or rising) and what moves mortgage rates

All three tools are read-only, declare an output schema and return structured content alongside the text. Class shares work in either spelling (BRK-B or BRK.B); leveraged and inverse funds never read above Hold.

Installation

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

claude.ai and the Claude desktop app: Settings, Connectors, add a custom connector with the URL and no authentication. ChatGPT: Plugins, then Add custom MCP server with the same URL. Gemini CLI users can install the vendor's extension.

Business Relevance

  • Operators with real-estate exposure: a keyless daily mortgage-rate read (Treasury direction, MBS momentum, environment) in whichever assistant you already use.
  • Investors and analysts: tiered ratings with factor-level breakdowns and a published grading record, for quick screens before deeper diligence.
  • Advisors and finance teams: consistent, sourced, dated rating context for client conversations, with methodology links in every result.

Integration with CorpusIQ

CorpusIQ keeps your own numbers consistent across AI clients - revenue, customers, cash. Market Eyes adds the outside view: ask your assistant what the rate environment looks like and how a public competitor is rated, in the same conversation where CorpusIQ reads your books.

Limitations

  • U.S.-listed stocks and ETFs only; crypto, futures and non-U.S. listings return a not-covered result, and five symbols shared with futures or coins (CORN, GOLD, WTI, BTC, ETH) are deliberately not rated.
  • Ratings are an informational product, not advice; every result carries a source and a disclaimer.
  • Brand new listing - no third-party track record yet; the vendor's own validation feed is public and machine-readable at marketeyeslive.com/api/validation-status.

FAQ

Is this investment advice?

No. It is a ratings product: conviction tiers and factor scores with sources and disclaimers, and a public record of how published ratings performed.

Does it need an API key?

No - the server is free and keyless, with stateless responses.

Which tickers are covered?

More than 11,000 U.S.-listed stocks and ETFs. Class shares work either way; tickers outside coverage return a result that says so instead of guessing.

How is a rating built?

A tier plus a 0 to 100 composite from eight factor scores (valuation, quality, momentum, earnings, sentiment, catalyst, risk-adjusted, macro fit). The methodology page and the grading feed are both public.

See Also

AI answers on your data. See where every number comes from.

Connect your tools and ask in ChatGPT, Claude, or Perplexity. Enterprise-grade answers in minutes. From $29.95/month.

Start free, no credit card
Contact Terms Privacy Pricing