[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79DARA6C9DT0XSPREN3DC2B":3,"skill-same:01M79DARA6C9DT0XSPREN3DC2B":49},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":34,"install":35,"resources":48},"01M79DARA6C9DT0XSPREN3DC2B","ml-best-practices",null,"CRITICAL RULE: You MUST use this skill whenever the task involves any machine learning tasks or data analysis.\nUse this skill if the user's prompt or requirements mention any of the following:\n  * Clustering\n  * Classification\n  * Regression\n  * Time series forecasting\n  * Statistical testing\n  * Model comparison\n  * ML\n  * Data analysis\n\nSQL\u002FBigQuery ML HANDOFF: If the user requires a SQL solution, use this skill to dictate the ANALYSIS STEPS (e.g., markdown analysis cells, visualization logic), but defer to `bigquery` for all SQL syntax.","NguyenPhuocAnhDung\u002Foil_forecast_tail_risk","NguyenPhuocAnhDung","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F167469630?v=4","01M799ZYV3KJAR2SVWBJ6HS2KT","Nguyễn Phước Anh Dũng","01M799ZYVSZ10CRDCF7JS9SA8C",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fworker_msB_fix\u002Fskills\u002Fml-best-practices\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: ml-best-practices\ndescription: |\n  CRITICAL RULE: You MUST use this skill whenever the task involves any machine learning tasks or data analysis.\n  Use this skill if the user's prompt or requirements mention any of the following:\n    * Clustering\n    * Classification\n    * Regression\n    * Time series forecasting\n    * Statistical testing\n    * Model comparison\n    * ML\n    * Data analysis\n\n  SQL\u002FBigQuery ML HANDOFF: If the user requires a SQL solution, use this skill to dictate the ANALYSIS STEPS (e.g., markdown analysis cells, visualization logic), but defer to `bigquery` for all SQL syntax.\nlicense: Apache-2.0\nmetadata:\n  version: v1\n  publisher: google\n---\n\n# ML Best Practices\n\nI want to read a story about the data, not just run code. Ensure every code cell\nis followed by a markdown cell analyzing the results. End the notebook with a\nsummary comprehensively answering the prompt.\n\nIf there is a good match between the user's request and a corresponding example\nplan, then adapt the example plan to fully answer the user's request:\n\n## Clustering:\n\nIdentify distinct groups based on their features.\n\n-   Understand the schema and field descriptions.\n-   Visualize features referenced in the prompt (e.g., with histograms,\n    scatterplots).\n-   Transform dates into timestamps.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Prefer to keep data instead of dropping it when possible.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder.\n-   Standardize numerical features.\n-   Perform clustering with a range of values, and collect the silhouette score.\n-   Choose the optimal number of clusters based on the silhouette score.\n-   Use dimensionality reduction (e.g., PCA) to project the data into two\n    dimensions.\n-   Scatterplot the samples in two dimensions with cluster labels as the hue.\n-   Scatterplot the samples in two dimensions with a discrete feature as the\n    hue.\n-   Describe the clusters in text by feature distributions or typical feature\n    values.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Time Series Forecasting:\n\nDevelop a predictive model to estimate future values based on historical trends.\nHow might different modeling approaches impact the prediction accuracy?\n\n-   Understand the schema and field descriptions.\n-   Visualize the target feature over time at a reasonable granularity.\n-   Always perform a chronological split on the data to create training,\n    validation, and test sets.\n-   Are there seasonal trends?\n-   Test for stationarity.\n-   Discuss possible modeling approaches. How might different modeling\n    approaches impact the prediction accuracy?\n-   Train two time series forecasting models to predict the target feature. Use\n    previous seasonality and stationarity information as model hyperparameters.\n-   Predict the target feature for the training and validation sets.\n-   Optionally, hypertune models with the validation set.\n-   Visualize the actual and predicted target feature vs time for each model on\n    the training and validation sets.\n-   Evaluate the validation performance with error metrics.\n-   Select a model.\n-   Retrain the selected model on the test and validation sets.\n-   Predict the test values with the selected model.\n-   Visualize the average target feature and the predicted test values.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Exploratory Data Analysis \u002F Anomaly Detection:\n\nIdentify and describe any outliers, unusual patterns, or significant trends\nobserved in the data. Provide visualizations to support your findings.\n\n-   Understand the schema and field descriptions.\n-   Visualize the target feature distribution in a way that shows outliers.\n-   Identify and describe any outliers in the target feature.\n-   Visualize relationships between the target feature and other features.\n-   Identify and describe unusual patterns or significant trends.\n-   Visualize patterns and trends.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Classification:\n\nGiven the data, can we classify by the target feature?\n\n-   Understand the schema and field descriptions.\n-   Identify rows that don't make sense. How many are there and what do they\n    contain?\n-   Identify rows without a target value. How many are there and what do they\n    contain?\n-   Drop rows that don't match the schema or don't have the target value (if it\n    is reasonable to do so).\n-   Split data into training, validation, and test sets.\n-   Create features to represent when data are missing, if this is meaningful.\n-   Handle missing data. Prefer to keep data instead of dropping it when\n    possible.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder.\n-   Standardize numerical features.\n-   Train multiple models.\n-   If there is evidence of overfitting, regularize and retrain the model.\n-   If there is evidence of underfitting, consider adding or engineering\n    features.\n-   Evaluate the models.\n-   Create confusion matrices.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Regression:\n\nPredict the continuous valued target feature.\n\n-   Understand the schema and field descriptions.\n-   Identify rows that don't make sense. How many are there and what do they\n    contain?\n-   Identify rows without a target value. How many are there and what do they\n    contain?\n-   Develop an understanding of the data and determine how to handle missing\n    values. This should make sense in the business context.\n-   Identify any potential sources of group leakage. Aggregate where appropriate\n    to prevent this.\n-   Visualize target feature.\n-   Split data into training, validation, and test sets.\n-   Handle missing data. Prefer to keep data instead of dropping it when\n    possible.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder. Restrict high cardinality\n    categorical features to a tractable size.\n-   Standardize numerical features.\n-   Train multiple models.\n-   Visualize the actual vs predicted values on training and validation data.\n-   If there is evidence of overfitting, regularize and retrain the model.\n-   If there is evidence of underfitting, consider adding or engineering\n    features.\n-   Evaluate the model error.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Comparing ML Models:\n\nEvaluate and compare multiple models to determine which is most suitable for\nproduction based on predictive power, robustness, and viability.\n\n-   Understand the schema and align metrics with business goals (e.g., cost of\n    false positives vs. false negatives).\n-   Establish baselines: define a naive baseline (majority class\u002Fmean) and a\n    simple ML baseline (e.g., Logistic\u002FLinear Regression).\n-   Ensure rigorous validation: use identical, fixed data splits for all models\n    and perform $k$-fold cross-validation.\n-   If data is temporal, use chronological splits for validation.\n-   Select and report metrics beyond accuracy (e.g., F1-Score, PR-AUC, MAE,\n    RMSE) that reflect business impact.\n-   Use bootstrapping to calculate 95% confidence intervals for key metrics to\n    determine statistical significance.\n-   Perform slice-based error analysis: evaluate model performance across key\n    subpopulations and demographics to identify bias or specific failure modes.\n-   Inspect and compare confusion matrices, residual plots, and calibration\n    curves.\n-   Evaluate operational trade-offs: consider inference latency, training time,\n    compute cost, and model size.\n-   Assess interpretability using tools like SHAP or LIME where transparency is\n    required.\n-   Conclusion: Recommend the optimal model for the specific use case,\n    justifying the choice with both performance and production viability.\n\n## No match:\n\n-   Understand the schema and field descriptions.\n-   Identify rows that don't make sense. How many are there and what do they\n    contain?\n-   Identify rows without a target value. How many are there and what do they\n    contain?\n-   Drop rows that don't match the schema or don't have the target value (if it\n    is reasonable to do so).\n-   Create features to represent when data are missing, if this is meaningful.\n-   Handle missing data. Prefer to keep data instead of dropping it when\n    possible.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder.\n-   Standardize numerical features.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Essential ML Practices\n\n[!IMPORTANT] ALWAYS follow these ML practices\n\n-   **Strict Featurization Ordering**: For supervised learning **ALWAYS** split\n    the dataset into training and test data **BEFORE** fitting preprocessing\n    pipelines (e.g. scaling, encoding). Fit the pipelines on the training data\n    and test data independently.\n\n-   **Handling Missing or NULL Values**: **ALWAYS** check for and handle missing\n    and NULL values. First, analyze their frequency. Then, decide whether to\n    keep them, drop them or impute them with a contextually appropriate value,\n    and explain your reasoning.\n","# ML Best Practices\n\nI want to read a story about the data, not just run code. Ensure every code cell\nis followed by a markdown cell analyzing the results. End the notebook with a\nsummary comprehensively answering the prompt.\n\nIf there is a good match between the user's request and a corresponding example\nplan, then adapt the example plan to fully answer the user's request:\n\n## Clustering:\n\nIdentify distinct groups based on their features.\n\n-   Understand the schema and field descriptions.\n-   Visualize features referenced in the prompt (e.g., with histograms,\n    scatterplots).\n-   Transform dates into timestamps.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Prefer to keep data instead of dropping it when possible.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder.\n-   Standardize numerical features.\n-   Perform clustering with a range of values, and collect the silhouette score.\n-   Choose the optimal number of clusters based on the silhouette score.\n-   Use dimensionality reduction (e.g., PCA) to project the data into two\n    dimensions.\n-   Scatterplot the samples in two dimensions with cluster labels as the hue.\n-   Scatterplot the samples in two dimensions with a discrete feature as the\n    hue.\n-   Describe the clusters in text by feature distributions or typical feature\n    values.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Time Series Forecasting:\n\nDevelop a predictive model to estimate future values based on historical trends.\nHow might different modeling approaches impact the prediction accuracy?\n\n-   Understand the schema and field descriptions.\n-   Visualize the target feature over time at a reasonable granularity.\n-   Always perform a chronological split on the data to create training,\n    validation, and test sets.\n-   Are there seasonal trends?\n-   Test for stationarity.\n-   Discuss possible modeling approaches. How might different modeling\n    approaches impact the prediction accuracy?\n-   Train two time series forecasting models to predict the target feature. Use\n    previous seasonality and stationarity information as model hyperparameters.\n-   Predict the target feature for the training and validation sets.\n-   Optionally, hypertune models with the validation set.\n-   Visualize the actual and predicted target feature vs time for each model on\n    the training and validation sets.\n-   Evaluate the validation performance with error metrics.\n-   Select a model.\n-   Retrain the selected model on the test and validation sets.\n-   Predict the test values with the selected model.\n-   Visualize the average target feature and the predicted test values.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Exploratory Data Analysis \u002F Anomaly Detection:\n\nIdentify and describe any outliers, unusual patterns, or significant trends\nobserved in the data. Provide visualizations to support your findings.\n\n-   Understand the schema and field descriptions.\n-   Visualize the target feature distribution in a way that shows outliers.\n-   Identify and describe any outliers in the target feature.\n-   Visualize relationships between the target feature and other features.\n-   Identify and describe unusual patterns or significant trends.\n-   Visualize patterns and trends.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Classification:\n\nGiven the data, can we classify by the target feature?\n\n-   Understand the schema and field descriptions.\n-   Identify rows that don't make sense. How many are there and what do they\n    contain?\n-   Identify rows without a target value. How many are there and what do they\n    contain?\n-   Drop rows that don't match the schema or don't have the target value (if it\n    is reasonable to do so).\n-   Split data into training, validation, and test sets.\n-   Create features to represent when data are missing, if this is meaningful.\n-   Handle missing data. Prefer to keep data instead of dropping it when\n    possible.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder.\n-   Standardize numerical features.\n-   Train multiple models.\n-   If there is evidence of overfitting, regularize and retrain the model.\n-   If there is evidence of underfitting, consider adding or engineering\n    features.\n-   Evaluate the models.\n-   Create confusion matrices.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Regression:\n\nPredict the continuous valued target feature.\n\n-   Understand the schema and field descriptions.\n-   Identify rows that don't make sense. How many are there and what do they\n    contain?\n-   Identify rows without a target value. How many are there and what do they\n    contain?\n-   Develop an understanding of the data and determine how to handle missing\n    values. This should make sense in the business context.\n-   Identify any potential sources of group leakage. Aggregate where appropriate\n    to prevent this.\n-   Visualize target feature.\n-   Split data into training, validation, and test sets.\n-   Handle missing data. Prefer to keep data instead of dropping it when\n    possible.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder. Restrict high cardinality\n    categorical features to a tractable size.\n-   Standardize numerical features.\n-   Train multiple models.\n-   Visualize the actual vs predicted values on training and validation data.\n-   If there is evidence of overfitting, regularize and retrain the model.\n-   If there is evidence of underfitting, consider adding or engineering\n    features.\n-   Evaluate the model error.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Comparing ML Models:\n\nEvaluate and compare multiple models to determine which is most suitable for\nproduction based on predictive power, robustness, and viability.\n\n-   Understand the schema and align metrics with business goals (e.g., cost of\n    false positives vs. false negatives).\n-   Establish baselines: define a naive baseline (majority class\u002Fmean) and a\n    simple ML baseline (e.g., Logistic\u002FLinear Regression).\n-   Ensure rigorous validation: use identical, fixed data splits for all models\n    and perform $k$-fold cross-validation.\n-   If data is temporal, use chronological splits for validation.\n-   Select and report metrics beyond accuracy (e.g., F1-Score, PR-AUC, MAE,\n    RMSE) that reflect business impact.\n-   Use bootstrapping to calculate 95% confidence intervals for key metrics to\n    determine statistical significance.\n-   Perform slice-based error analysis: evaluate model performance across key\n    subpopulations and demographics to identify bias or specific failure modes.\n-   Inspect and compare confusion matrices, residual plots, and calibration\n    curves.\n-   Evaluate operational trade-offs: consider inference latency, training time,\n    compute cost, and model size.\n-   Assess interpretability using tools like SHAP or LIME where transparency is\n    required.\n-   Conclusion: Recommend the optimal model for the specific use case,\n    justifying the choice with both performance and production viability.\n\n## No match:\n\n-   Understand the schema and field descriptions.\n-   Identify rows that don't make sense. How many are there and what do they\n    contain?\n-   Identify rows without a target value. How many are there and what do they\n    contain?\n-   Drop rows that don't match the schema or don't have the target value (if it\n    is reasonable to do so).\n-   Create features to represent when data are missing, if this is meaningful.\n-   Handle missing data. Prefer to keep data instead of dropping it when\n    possible.\n-   Before applying encoders, check if the dataset already contains pre-encoded\n    features and prefer existing numerical representations.\n-   Transform ordinal data with an ordinal encoder.\n-   Transform nominal data with a one hot encoder.\n-   Standardize numerical features.\n-   Conclusion: comprehensively answer the prompt in a final markdown cell.\n\n## Essential ML Practices\n\n[!IMPORTANT] ALWAYS follow these ML practices\n\n-   **Strict Featurization Ordering**: For supervised learning **ALWAYS** split\n    the dataset into training and test data **BEFORE** fitting preprocessing\n    pipelines (e.g. scaling, encoding). Fit the pipelines on the training data\n    and test data independently.\n\n-   **Handling Missing or NULL Values**: **ALWAYS** check for and handle missing\n    and NULL values. First, analyze their frequency. Then, decide whether to\n    keep them, drop them or impute them with a contextually appropriate value,\n    and explain your reasoning.\n",{"name":5,"license":15,"metadata":24,"description":27},{"version":25,"publisher":26},"v1","google","CRITICAL RULE: You MUST use this skill whenever the task involves any machine learning tasks or data analysis.\nUse this skill if the user's prompt or requirements mention any of the following:\n  * Clustering\n  * Classification\n  * Regression\n  * Time series forecasting\n  * Statistical testing\n  * Model comparison\n  * ML\n  * Data analysis\n\nSQL\u002FBigQuery ML HANDOFF: If the user requires a SQL solution, use this skill to dictate the ANALYSIS STEPS (e.g., markdown analysis cells, visualization logic), but defer to `bigquery` for all SQL syntax.\n","main","89920776f7cbf648024818498d24874504a8d781",".agents\u002Fworker_msB_fix\u002Fskills\u002Fml-best-practices",1,"https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung","2026-09-24 13:39:29",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add NguyenPhuocAnhDung\u002Foil_forecast_tail_risk --skill ml-best-practices",{"tool":41,"command":42},"bunx","bunx skills add NguyenPhuocAnhDung\u002Foil_forecast_tail_risk --skill ml-best-practices",{"tool":44,"command":45},"pnpm","pnpm dlx skills add NguyenPhuocAnhDung\u002Foil_forecast_tail_risk --skill ml-best-practices","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fworker_msB_fix\u002Fskills\u002Fml-best-practices\u002FSKILL.md",true,[],[50,58,65,72,79,86],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":19},"01M79D3735P38W4M5PPKY8V7ZN","verification-loop","A comprehensive verification system for Claude Code sessions.","MIT","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fecc\u002Fskills\u002Fverification-loop\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M79CYA4HEQN5ZFD0TD5YKW6J","returns-reverse-logistics","Codified expertise for returns authorization, receipt and inspection, disposition decisions, refund processing, fraud detection, and warranty claims management. Informed by returns operations managers with 15+ years experience. Includes grading frameworks, disposition economics, fraud pattern recognition, and vendor recovery processes. Use when handling product returns, reverse logistics, refund decisions, return fraud detection, or warranty claims.","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fecc\u002Fskills\u002Freturns-reverse-logistics\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M79CXQ5Y3TGM7WV8B4MR7W8A","recursive-decision-ledger","Use when the user asks for repeated rollouts, marked decision processes, high-dimensional search, stochastic optimization, local-optima exploration, ensemble comparison, or recursive reasoning with a visible evidence trail.","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fecc\u002Fskills\u002Frecursive-decision-ledger\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M79CW442W74WMRYBQS3278T9","quality-nonconformance","Codified expertise for quality control, non-conformance investigation, root cause analysis, corrective action, and supplier quality management in regulated manufacturing. Informed by quality engineers with 15+ years experience across FDA, IATF 16949, and AS9100 environments. Includes NCR lifecycle management, CAPA systems, SPC interpretation, and audit methodology. Use when investigating non-conformances, performing root cause analysis, managing CAPAs, interpreting SPC data, or handling supplier quality issues.","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fecc\u002Fskills\u002Fquality-nonconformance\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M79CV8N0CTPG6XXJ5WQNC6YE","production-scheduling","Codified expertise for production scheduling, job sequencing, line balancing, changeover optimization, and bottleneck resolution in discrete and batch manufacturing. Informed by production schedulers with 15+ years experience. Includes TOC\u002Fdrum-buffer-rope, SMED, OEE analysis, disruption response frameworks, and ERP\u002FMES interaction patterns. Use when scheduling production, resolving bottlenecks, optimizing changeovers, responding to disruptions, or balancing manufacturing lines.","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fecc\u002Fskills\u002Fproduction-scheduling\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M79CSS7SSDH2846T9JPVNNX6","parallel-execution-optimizer","Use when the user wants a task done much faster through parallel work, concurrent agents, batched tool calls, isolated worktrees, or many independent verification lanes without losing correctness.","https:\u002F\u002Fgithub.com\u002FNguyenPhuocAnhDung\u002Foil_forecast_tail_risk\u002Fblob\u002Fmain\u002F.agents\u002Fecc\u002Fskills\u002Fparallel-execution-optimizer\u002FSKILL.md",[],[]]