Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    HPSILab Quant Research — Stocks & Options

    0

    Let your agent run IV, options pressure, Monte Carlo and AI-signal research on US stocks via the HPSILab MCP, and report gaps honestly.

    Free

    3 usesSecurity scanned
    HPSILab Quant Research — Stocks & Options

    HPSILab Quant Research — Stocks & Options

    Example session with this skill installed

    Give me a quick quant read on NVDA — volatility, options positioning and where price might go over the next two weeks.

    • Read your context and instructions
    • Compiled the hpsilab quant research

    NVDA — quant snapshot (data as of the 2026-10-15 close)

    Overall: All 5 underlying modules returned OK (complete result). HPSILab's direction model screens NVDA

    bullish, but with a modest 44/100 direction score: most signals lean up, while options skew still shows demand for downside protection.

    Volatility (IV Radar)

    MetricValueRead
    ATM IV35.9%Moderate for NVDA
    30-day IV36.7%—
    IV rank / percentile56.25 / 44.16Mid-range vs. the past year
    25-delta risk reversal−3.5 vol ptsPut-heavy: traders pay up for downside
    RegimeNeutralNo strong options bias

    Options positioning (expiry 2026-11-13)

    LevelPrice
    Spot$217.23
    Max Pain$230.00 (above spot)
    Gamma Wall / Put Wall$200.00 (support zone)
    Call Wall$240.00 (resistance zone)
    Likely weekly high$230.00
    Likely expiry close range$225–$235
    Weekly expected move±$10.40

    Next ~10 trading days (Monte Carlo)

    • Median projected close $218.34 (mean $219.02)
    • 90% of simulated paths land between $192.55 and $247.11
    • Only 7.5% of paths finish above the $243.37 resistance

    AI next-day signal

    • Probability of an up close:

    60.1%, but model consensus is

    mixed (Random Forest 70.4% up, Logistic Regression 49.7%), so treat this as a weak lean, not a call.

    Strategy backtest (momentum long on NVDA)

    Total return 54.2%, but Sharpe only 0.24 with a −24.9% max drawdown and a 37.6% win rate — returns came with meaningful risk.

    What this adds up to

    • Observed data: spot $217, Max Pain $230, put-side gamma support near $200, put-heavy skew.
    • Model output: slight upward lean (60% AI probability, median MC close ≈ spot), with a wide ±$27 range.

    Inference: options positioning points to a pull toward $225–$235 into the November expiry, but skew and the mixed AI consensus say the move is far from certain.

    Next steps you can ask for: a deeper IV term-structure view, the chart pack, or a pre-trade risk scan (Pro tool; may use Credits).

    For research only. Not investment advice.

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

    About this skill

    HPSILab Quant Research teaches your agent how to use the HPSILab MCP server for quantitative research on US stocks and options — which tool to call for each question, how to tell a real success from a partial or failed result, and how to write the answer without overstating what the models say.

    What your agent can do with it

    • One-call stock overview combining implied volatility, options positioning, a Monte Carlo range and an AI direction signal, with bullish and bearish factors
    • Implied volatility: ATM IV, 30-day IV, IV rank/percentile, 25-delta skew, squeeze score
    • Options pressure map: Max Pain, dealer Gamma Wall, call/put walls, likely weekly high, expected expiry range
    • Monte Carlo: ~10-trading-day price range and odds of finishing above resistance
    • AI next-day prediction: probability of an up close and model consensus
    • Backtest metrics: total return, Sharpe, max drawdown, win rate
    • Charts and full research reports (Pro tools on HPSILab)

    Why use the skill instead of calling the tools raw

    • Right tool, fewer calls: overview first, dedicated tools only for detail or for modules that came back missing — no repeated full analyses burning your quota.
    • Honest results: the agent checks isError, status, complete, warnings and every sub-tool response, labels partial results and never fills gaps with guessed numbers.
    • Clear error handling: auth, rate limits, plan limits, payment-required and bad tickers each get a specific, non-looping response.
    • Safe by design: research only, no trade execution, no personalized advice, never asks for or prints API keys, and never follows checkout or upgrade links from server responses.

    Requirements

    This is an instructions-only skill (no scripts, no network calls of its own). It needs the HPSILab MCP server connected in your agent (https://hpsilab.com/mcp) with your own HPSILab API key. HPSILab usage, plans and Credits are not included with this skill.

    For research only. Not investment advice.

    US stocks and options only. Requires HPSILab API key and credits for Pro tools. No execution or broker integration.

    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 any payment.

    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

    3 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    HPSILab MCP
    HPSILab account and API key

    Permissions

    Network Access
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    A standard agent has no live options or volatility data and may guess. With this skill, your agent pulls IV, options positioning, Monte Carlo ranges and AI probabilities from the HPSILab MCP tools, checks whether each result is complete, and labels partial results instead of filling gaps. It is research only and does not give personalized investment advice.

    The skill is designed for any agent environment that supports the Model Context Protocol (MCP), though it requires a valid HPSILab API key and connection to their hosted server to function.

    Currently, the skill is specialized for the US equity and options markets; it does not support international exchanges, forex, or cryptocurrency at this time.

    The skill is free. You get the SKILL.md instructions (plus README and MIT license) that teach your agent how to use the HPSILab MCP tools. HPSILab itself is a separate service: API usage, plans and Credits are not included and are governed by your own HPSILab account.

    Setup is straightforward: you link the HPSILab MCP server address to your agent environment and input your HPSILab API key to begin calling tools like 'analyze_stock' and 'get_iv_radar'.

    Both. Most tools return structured text and JSON. 'generate_stock_images' returns chart image URLs (IV radar, option pressure, Monte Carlo, AI prediction, equity curves), and the Pro tool 'generate_stock_research_report' returns a markdown report with five embedded charts.

    New in Trading & Investing