EP Ratio Screener Server

Overview

The EP Ratio Screener Server runs a fundamental stock screener based on normalized earnings yield, balance-sheet health, ROIC, earnings consistency, and yield spread versus the risk-free rate.

Connection

Add this server to your MCP client configuration.

Cursor
Bearer auth

{
  "mcpServers": {
    "ep-ratio-screener": {
      "url": "https://ep-ratio-screener-production.up.railway.app/mcp/epratioscreener"
    }
  }
}

{
  "mcpServers": {
    "ep-ratio-screener": {
      "url": "https://ep-ratio-screener-production.up.railway.app/mcp/epratioscreener",
      "headers": {
        "Authorization": "Bearer ${env:EP_RATIO_SCREENER_MCP_TOKEN}"
      }
    }
  }
}

Restart the client after changing MCP configuration. The server tools appear automatically after the connection is established.

Transport

Property	Value
Protocol	MCP over Streamable HTTP
MCP URL	`https://ep-ratio-screener-production.up.railway.app/mcp/epratioscreener`
Health URL	`https://ep-ratio-screener-production.up.railway.app/health`
MCP path	`/mcp/epratioscreener`
Request envelope	`{"request": {...}}`
Auth	Optional bearer token when enabled for the endpoint

Best-Fit Workflows

Validate runtime configuration before screening.
Score one ticker with the E/P-based decision pipeline.
Screen a batch of tickers and export CSV outputs.
Check whether the service is running in full-data or fallback mode.

Recommended Tools

config-validate
scoring-engine-score
screen-tickers

Tools

Tool	Description	Returns
`config-validate`	Validates the current EPRatio Screener configuration. Params: none	`Any`
`scoring-engine-score`	Scores a ticker using the EPRatio screener scoring engine. Params: ticker (str): ticker symbol to score	`Any`
`screen-tickers`	Run the EPRatio screening workflow and save results as CSV files. Outputs a full results CSV and a filtered passing CSV sorted by yield spread. Params: tickers: optional list of ticker symbols; use_fmp_screen: if True, overrides tickers and uses FMP’s screened universe	`dict[str, Any]`

Examples

Validate config

{
  "request": {}
}

Score one ticker

{
  "request": {
    "ticker": "MSFT"
  }
}

Screen a batch

{
  "request": {
    "tickers": [
      "MSFT",
      "AAPL",
      "JPM"
    ],
    "use_fmp_screen": false
  }
}

n8n Workflow

This server can be called from an n8n schedule or POST Webhook to run a ticker screen and return a ranked JSON report. Recommended n8n HTTP Request chain:

Accept tickers and use_fmp_screen from the webhook body, or use a scheduled default universe.
Initialize the MCP session and store MCP-Session-Id.
Call load_skills once when the endpoint requires skill bootstrap.
Split larger ticker lists into batches.
Call screen-tickers with {"request":{"tickers":[...],"use_fmp_screen":false}}.
Parse the SSE response, combine batch results, and route the output to Respond to Webhook or delivery nodes.

Use n8n Credentials for any delivery channel. Do not export private webhook paths, email addresses, or OAuth credential references with workflow JSON. See n8n Workflows for the shared webhook and MCP session patterns. Available n8n template:

EP Ratio Screener JSON - public ticker-screen webhook at POST https://n8n.limex.pro/webhook/a1b2c3d4-e5f6-7890-abcd-ef1234567890. The template removes the scheduled email-delivery branch; add your own delivery nodes after import.

Notes

use_fmp_screen=true requires the optional FMP data source to be configured.
Fallback mode still supports the core scoring flow, but has less universe/enrichment coverage.

Client setup

Configure this endpoint in Cursor, Claude Desktop, or a generic MCP client.

n8n workflows

Build webhook and scheduled screening flows around this MCP endpoint.

Shared tools

Use health, result, artifact, environment, and table helper tools.

Other Servers

Dataloader

Market-data retrieval, TA-Lib indicators, and dataframe exports.

Fama-French Replicate

Official and replicated Fama-French factors plus loadings and alpha estimation.

Statistical Factor Models

Stock-Watson, complete-panel, and dynamic statistical factor extraction.

Jump Models

JumpModel and SparseJumpModel regime fitting, online prediction, and backtesting.

Wavelet Mean Reversion

Wavelet-based mean reversion analysis for financial time series.

Parallax ExtremeHurst

ExtremeHurst signal generation from OHLCV data.

Volatility Scaling Lab

Volatility targeting, EWMA volatility, Monte Carlo bands, and risk diagnostics.

Overview

MCP Setup

Dataloader Server

Fama-French Replicate Server

Statistical Factor Models Server

Jump Models Server

Wavelet Mean Reversion Server

Parallax ExtremeHurst Server