More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ML for Risk, Fraud and Revenue

    1

    Detect fraud and anomalies, score credit, set prices and predict churn and failures. 10 ordered skills, starting with Anomaly Detection, Fraud Detection, Credit Scoring and Credit Risk.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    ML for Risk, Fraud and Revenue

    ML for Risk, Fraud and Revenue

    Example session with this skill installed

    Find anomalies in this machine sensor log. We have 10,000 rows of pressure and temperature data. The team can only check 5 alerts per day. Use Isolation Forest.

    • Read your context and instructions
    • Compiled the ml for risk

    I have analyzed the sensor log and designed an anomaly detection pipeline focused on your alert budget of 5 per day.

    Anomaly Detection Plan: Machine Sensor Log

    StepActionDetail
    1. FramingDetector ChoiceIsolation Forest (unsupervised)
    2. PrepScalingRobustScaler to handle heavy-tailed pressure spikes
    3. BudgetContaminationSet to 0.0005 (5 alerts / 10,000 rows)
    4. FeaturesEngineeringCreated rolling 1h mean and lag differences for temperature
    5. ValidationProxy MetricCalculated score stability and feature importance (SHAP)
    import pandas as pd
    from sklearn.ensemble import IsolationForest
    from sklearn.preprocessing import RobustScaler
    
    # Load data
    df = pd.read_csv('sensor_log.csv')
    features = ['pressure', 'temperature', 'temp_roll_mean', 'press_diff']
    
    # Scale and fit
    scaler = RobustScaler()
    X = scaler.fit_transform(df[features])
    
    # Fit Isolation Forest with specific contamination for alert budget
    model = IsolationForest(contamination=0.0005, random_state=42)
    df['anomaly_score'] = model.fit_predict(X)
    
    # Flag top 5 records
    anomalies = df[df['anomaly_score'] == -1]
    

    Next steps

    • Review the 5 flagged records for contextual patterns (e.g., time of day).
    • If these are false positives, increase the rolling window features to capture longer trends.
    • Check for sensor drift by comparing the first and last 1,000 rows of the score distribution.

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

    About this skill

    The problem

    Fintechs, retail and industry often need a repeatable process, not a one-off prompt. This pack targets: Detect fraud and anomalies, score credit, set prices and predict churn and failures.

    What's inside (10 skills, suggested learning order)

    1. Anomaly Detection — Identify fraud and failures.
    2. Fraud Detection — Anti-fraud models.
    3. Credit Scoring and Credit Risk — Model default.
    4. Financial Risk Modeling — Quantify market risks.
    5. Quantitative Financial Analysis — Models for investments.
    6. Dynamic Pricing and Price Optimization — Set prices with models.
    7. Churn and Purchase Propensity — Propensity models.
    8. Predictive Lead Scoring — Score leads by probability.
    9. Predictive Maintenance — Predict equipment failures.
    10. Survival Analysis and Time-to-Event — Model churn and failures over time.

    Why a skill pack instead of a one-off prompt

    Each skill gives the agent a defined trigger, step-by-step procedure and output format, so results are consistent across sessions and easier to review.

    How to use

    1. Install the pack in your SKILL.md-compatible agent.
    2. Ask for the task in plain language.
    3. Review the output against the skill's quality checks before using it.

    Known limitations

    • Outputs are drafts that need human review; they are not professional, legal or financial advice.
    • Tools, APIs and prices change; check current vendor documentation before relying on version-specific details.
    • Quality depends on the context you give the agent (data, requirements, constraints).
    • Tested only where stated in the Compatibility note.

    Also by Rodrigo Santos

    Specialized Agents and Distribution; AI Agents: Fundamentals and Protocols; Model Fine-Tuning and Alignment

    Requires historical event data with timestamps. Performance is unstable with very low conversion counts. Does not provide investment advice.

    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

    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 3 days ago

    Permissions

    No special permissions declared or detected

    Listed3 days ago

    What's inside

    Frequently Asked Questions

    This skill provides ten specialized workflows for high-stakes modeling, specifically designed to eliminate common errors like data leakage and look-ahead bias that generic AI prompts often overlook.

    The skill is built for Python-centric environments and is compatible with AI agents capable of executing code via libraries like pandas, scikit-learn, lifelines, and QuantLib.

    You receive a collection of ten production-ready workflows covering anomaly detection, credit risk, propensity, pricing optimization, and backtesting, including all necessary preprocessing and validation steps.

    Yes, the risk and credit modeling components are built using industry-standard methodologies (like Gini and PD) that prioritize interpretability, making it easier to explain model decisions for regulatory compliance.

    The skill requires historical, timestamped event data to function; it is not suitable for cold-start scenarios where no prior failure, fraud, or transaction data exists.

    New in Finance & Accounting