More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Multi-Factor Stock Scorer: Value, Quality, Momentum

    1

    Your agent then explains the result under strict rules: every claim cites a printed number, missing data is named, and nothing is made up.

    $19

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Multi-Factor Stock Scorer: Value, Quality, Momentum

    Multi-Factor Stock Scorer: Value, Quality, Momentum

    Example session with this skill installed

    Rank Apple, Microsoft, Nvidia, Alphabet, Amazon, Meta, JPMorgan, Exxon, Johnson & Johnson and Walmart on value, quality, momentum, low volatility and growth. Which one comes out on top, and why?

    • Read your context and instructions
    • Compiled the multi-factor stock scorer

    Summary

    Of these 10 stocks,

    XOM ranks first with a score of 67, on Value 100 and Momentum 89, and

    WMT ranks last with 13, scoring 11 or lower on four of the five pillars (Value 11, Quality 0, Momentum 11, Growth 11). What separates first from last most is Value (100 against 11) and Momentum (89 against 11).

    Multi-factor scores · 10 stocks · 2026-11-05

    #TickerNameSectorScoreValueQualityMomentumLow volGrowthData
    1XOMExxonMobil Holdings Corporati…Energy█████▍·· 6710011896778100%
    2MSFTMicrosoft CorporationTechnology█████··· 634489673367100%
    3JPMJPMorgan Chase & Co.Financial Services████▉··· 61894456893377%
    4METAMeta Platforms, Inc.Communication Services████▌··· 57786744089100%
    5AAPLApple Inc.Technology████▍··· 550567810044100%
    6NVDANVIDIA CorporationTechnology████▎··· 53221002222100100%
    7JNJJohnson & JohnsonHealthcare████▏··· 523333100780100%
    8GOOGLAlphabet Inc.Communication Services████▏··· 516778331156100%
    9AMZNAmazon.com, Inc.Consumer Cyclical██▎····· 28562205622100%
    10WMTWalmart Inc.Consumer Defensive█▏······ 13110114411100%

    Score = weighted average of the pillar scores (value 20%, quality 25%, momentum 25%, low vol 15%, growth 15%). Pillar scores are percentiles across all the stocks given: 100 = best of this group, 0 = worst. Data = share of the 13 inputs available.

    The numbers behind the scores

    TickerP/EP/BEV/EBITDAROEOp. marginDebt/eq12m ex 1m6mVol 60dMax fall 1yRevenue g.Earnings g.
    XOM17.02.59.115.5%17.0%0.13x+47.9%+14.3%26.9%-20.1%+32.6%+111.4%
    MSFT28.08.520.434.2%49.8%0.28x+3.8%+27.9%26.8%-30.5%+18.8%+32.3%
    JPM14.22.5n/a18.0%42.6%n/a+8.6%+11.2%20.2%-15.5%+10.2%+9.9%
    META20.65.714.132.3%35.4%0.42x+15.4%-1.7%44.6%-28.7%+26.1%+264.8%
    AAPL38.344.629.8142.3%31.1%0.84x+25.1%+19.9%22.1%-13.8%+11.7%+14.1%
    NVDA23.019.223.8117.2%66.2%0.17x+19.8%-13.9%39.9%-24.1%+105.9%+127.8%
    JNJ41.38.020.120.0%25.7%0.55x+41.4%+23.7%25.1%-11.0%+5.9%-95.9%
    GOOGL14.85.318.644.3%31.8%0.22x+26.6%-28.9%35.6%-30.6%+24.7%-25.3%
    AMZN21.04.616.727.4%13.9%0.50x+4.3%-4.0%31.3%-20.5%+12.1%-3.1%
    WMT46.310.324.822.3%3.5%0.72x+6.6%-1.5%30.8%-23.3%+5.9%-9.1%

    Evidence cards

    1. XOM · ExxonMobil Holdings Corporation · score 67

    Strongest: Value 100 (P/E 17.0, P/B 2.5, EV/EBITDA 9.1); Momentum 89 (Return 12m excl. last month +47.9%, Return 6m +14.3%)

    • Weakest: Quality 11 (Return on equity 15.5%, Operating margin 17.0%, Gross margin 29.8%, Debt / equity 0.13x)

    2. MSFT · Microsoft Corporation · score 63

    Strongest: Quality 89 (Return on equity 34.2%, Operating margin 49.8%, Gross margin 67.7%, Debt / equity 0.28x); Momentum 67 (Return 12m excl. last month +3.8%, Return 6m +27.9%)

    • Weakest: Low vol 33 (Volatility 60d (annualised) 26.8%, Worst fall from a high (1y) -30.5%)

    3. JPM · JPMorgan Chase & Co. · score 61

    Strongest: Value 89 (P/E 14.2, P/B 2.5); Low vol 89 (Volatility 60d (annualised) 20.2%, Worst fall from a high (1y) -15.5%)

    • Weakest: Growth 33 (Revenue growth (latest quarter, yoy) +10.2%, Earnings growth (latest quarter, yoy) +9.9%)
    • Data not available for: EBITDA / EV, Gross margin, Debt / equity

    Source: Yahoo Finance (free, unofficial): latest trailing fundamentals as Yahoo reports them (it does not date each figure; growth is the latest quarter against a year earlier), daily adjusted closes 2024-11-06 to 2026-11-05, fetched 2026-11-05 21:42 UTC.

    What drives the ranking

    XOM leads on cheapness and trend: the lowest EV/EBITDA here (9.1) and the highest return over 12 months excluding the last (+47.9%). Its weak spot is Quality 11, with an operating margin of 17.0% against MSFT's 49.8%.

    MSFT is second on Quality 89 (operating margin 49.8%, ROE 34.2%). Its Low vol score of 33 reflects a -30.5% worst fall from a high over the year.

    • JPM is third on Value 89 (P/E 14.2, the lowest here) and Low vol 89 (60-day volatility 20.2%, the lowest here).

    NVDA tops Quality and Growth (100 each: operating margin 66.2%, revenue growth +105.9%) but scores 22 on Value, Momentum and Low vol, which puts it sixth at 53.

    AAPL has the best Low vol score (100: volatility 22.1%, worst fall -13.8%) and the worst Value score (0: P/E 38.3, P/B 44.6).

    Caveats

    • JPM is scored on 77% of the inputs. Data not available for EBITDA / EV, gross margin and debt / equity, which this data source does not report for banks, so its Value score rests on P/E and P/B alone.
    • Two growth figures look like one-offs: META's latest-quarter earnings growth of +264.8% and JNJ's -95.9%. JNJ's Growth score of 0 rests largely on that number.
    • The group mixes energy, banking, technology and retail, so margins are compared across very different businesses. Sector-neutral mode needs at least 4 stocks per sector to change anything.
    • Scores rank these 10 stocks against each other only.

    Next checks

    • Tilt toward value and quality: --weights value=0.4,quality=0.4,momentum=0.2, to see whether XOM keeps first place.
    • Add more banks and energy names and rerun with --sector-neutral to compare JPM and XOM with their own sectors.
    • Read JNJ's and META's latest quarterly reports for the earnings growth figures (-95.9% and +264.8%) before relying on the Growth pillar.
    • Keep the audit trail with --json evidence.json: every input, z-score and percentile behind these scores.

    For research and education. Not investment advice.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    What it does

    Give your agent a list of stocks and get a factor ranking you can audit: a score from 0 to 100 for each stock, five pillar scores, the raw numbers behind them, and evidence cards for the leaders. Your agent then explains the result under strict rules: every claim cites a printed number, missing data is named, and nothing is made up.

    • Five pillars, 13 inputs: Value (earnings yield, book yield, EBITDA/EV), Quality (ROE, operating and gross margin, debt/equity), Momentum (12 months excluding the last, 6 months), Low volatility (60-day volatility, worst fall over a year) and Growth (revenue and earnings, latest quarter on a year earlier)
    • Your weights: tilt toward value, quality or momentum in one flag
    • Sector-neutral mode: compare banks with banks and software with software
    • Evidence cards: strongest and weakest pillar of each leader, with the numbers that put them there
    • Your own data: a CSV from your terminal or spreadsheet, merged with Yahoo or used fully offline
    • Audit trail: --json writes every input, z-score and percentile; --save-csv writes a spreadsheet

    How it works

    The bundled script (scripts/score.py) fetches the latest fundamentals and two years of daily adjusted prices from Yahoo Finance, or reads your CSV. It then scores each input across the stocks you gave: winsorized at the 5th and 95th percentiles, turned into a median-based z-score capped at plus or minus 3, flipped where lower is better, averaged into pillars, ranked as percentiles, and combined with your weights. Missing inputs are never filled in: the weights are rescaled over the pillars a stock has, and stocks resting on too little data are flagged as evidence limited.

    The agent picks the settings from your request, runs the script, pastes its output as printed, and writes a summary, what drives the ranking, the caveats and concrete next checks, using only the printed numbers.

    Why you can trust it

    • A robust, standard method from institutional factor research, documented step by step in references/method.md.
    • Valuation is scored as yields, so loss-makers rank last on earnings instead of vanishing from the comparison.
    • Scores are explicitly relative: the output says "best of this group", never "cheap" in the absolute.
    • Every run ends with a source line: data provider, price dates and fetch time.
    • No buy or sell calls, price targets or trades, ever.

    Tested

    • 26 automated tests, all offline: the scoring maths against hand-worked values, the factor inputs, Yahoo's responses (including an expired session token), CSV percent handling, and the command end to end. All pass on Python 3.12 and 3.13.
    • Real runs on live Yahoo data for 10 and 15 US large caps, in plain and sector-neutral mode.
    • Run in Claude Code.

    Known limits

    • Yahoo's endpoints are unofficial: data can be delayed, requests rate-limited, and the service can change. In some regions Yahoo asks for cookie consent before issuing the session token the fundamentals need; the scorer then says so, and your own CSV works instead.
    • Fundamentals are the latest trailing figures, without filing dates: good for ranking today, not for backtesting.
    • No analyst estimates or revisions, no forecasts, no trading signals.
    • With fewer than 8 stocks, percentiles are coarse, and the output warns you.

    For research and education. Not investment advice.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Trading & Investing.

    Requires Python 3.10+. Yahoo's endpoints are unofficial: data can be delayed, requests rate-limited, and the service can change; in some regions Yahoo asks for cookie consent before issuing the session token the fundamentals need, and your own CSV then works instead. Fundamentals are the latest trailing figures without filing dates, so it ranks today and cannot backtest. No analyst estimates, forecasts or trading signals. Scores are relative to the stocks given.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 11 days ago

    Accounts
    Yahoo
    Software
    Python 3.10+

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access

    Allowed Hosts

    query1.finance.yahoo.com
    fc.yahoo.com
    finance.yahoo.com

    File Scopes

    multi-factor-stock-scorer/**
    **/*.csv
    **/*.json

    Network: read-only requests to finance.yahoo.com, fc.yahoo.com and query1.finance.yahoo.com with the ticker symbols; Yahoo's session cookie is kept in memory for the run only. None with --offline. Read Files: the CSV or ticker list you pass. Write Files: only the files you name with --json or --save-csv.

    Listed11 days ago

    What's inside

    Frequently Asked Questions

    No. It uses Yahoo Finance's free endpoints with no account. If you have your own data, for example a terminal export, pass it as a CSV and it is used instead of, or with, Yahoo's.

    That the stock beats about 80% of the other stocks you gave on the weighted pillars. Scores are relative to your list, never absolute: a list of expensive stocks still has a best value stock.

    Missing inputs are never filled in. The pillar uses the inputs it has, the weights are rescaled over the pillars present, and a stock with under half the inputs or fewer than 3 pillars is flagged as evidence limited.

    Yes: --weights value=0.4,quality=0.4,momentum=0.2 for any mix of the five pillars, and --sector-neutral to compare each stock with its own sector where 4 or more share it.

    Only read-only requests to Yahoo Finance with the ticker symbols. Yahoo's session cookie is kept in memory for the run and never saved. With --offline there is no network call at all.

    No. It ranks and explains with numbers, and the agent is instructed never to give buy or sell calls, price targets or position sizes, or to place orders.

    New in Trading & Investing