[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M73C289R52CQTPRC8G43VTNH":3,"skill-same:01M73C289R52CQTPRC8G43VTNH":65},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":37,"commitSha":38,"skillDirectory":39,"forksCount":40,"repositoryHtmlUrl":41,"repositoryHomepageUrl":6,"ownerHtmlUrl":42,"ownerBlogUrl":43,"ownerDescription":6,"updatedAt":44,"install":45,"resources":58},"01M73C289R52CQTPRC8G43VTNH","setup-automl-pipeline",null,"Configure automated ML pipelines using Optuna or Ray Tune for hyperparameter optimization. Implement efficient search strategies (Hyperband, ASHA), define search spaces, and set up early stopping to find optimal model configurations with minimal manual tuning. Use when starting a new ML project, retraining with new data and re-optimizing hyperparameters, comparing multiple algorithms, or when the team lacks deep expertise in specific algorithm hyperparameters.","roohe\u002Fagentic-super-skills","roohe","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99054084?v=4","01M734WCDFP6RZVT5CNM00QR26","01M734WCE4BD0T4NZS5PCV9V5F",10,"MIT","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fsetup-automl-pipeline\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: setup-automl-pipeline\nlocale: caveman-lite\nsource_locale: en\nsource_commit: 82c77053\ntranslator: \"Julius Brussee homage — caveman\"\ntranslation_date: \"2026-05-03\"\ndescription: >\n  Configure automated ML pipelines using Optuna or Ray Tune for hyperparameter\n  optimization. Implement efficient search strategies (Hyperband, ASHA), define\n  search spaces, and set up early stopping to find optimal model configurations\n  with minimal manual tuning. Use when starting a new ML project, retraining\n  with new data and re-optimizing hyperparameters, comparing multiple\n  algorithms, or when the team lacks deep expertise in specific algorithm\n  hyperparameters.\nlicense: MIT\nallowed-tools: Read Write Edit Bash Grep Glob\nmetadata:\n  author: Philipp Thoss\n  version: \"1.0\"\n  domain: mlops\n  complexity: intermediate\n  language: multi\n  tags: automl, optuna, ray-tune, hyperparameter, optimization, hyperband, asha\n---\n\n# Setup AutoML Pipeline\n\n\n> See [Extended Examples](references\u002FEXAMPLES.md) for complete configuration files and templates.\n\nAutomate hyperparameter tuning and model selection using Optuna or Ray Tune with efficient search strategies.\n\n## When to Use\n\n- Starting a new ML project and needing good model configurations quickly\n- Retraining an existing model with new data and re-optimizing hyperparameters\n- Comparing multiple algorithms and their optimal configurations\n- Limited time for manual tuning but need near-optimal performance\n- Team lacks deep expertise in specific algorithm hyperparameters\n- Need a reproducible and documented optimization process\n\n## Inputs\n\n- **Required**: Training dataset with features and labels\n- **Required**: Validation dataset for objective evaluation\n- **Required**: Model type(s) to optimize (e.g., XGBoost, LightGBM, neural network)\n- **Required**: Optimization objective (metric to maximize\u002Fminimize)\n- **Required**: Compute budget (time or number of trials)\n- **Optional**: Search space constraints (min\u002Fmax values for hyperparameters)\n- **Optional**: Prior knowledge of good hyperparameter ranges\n\n## Procedure\n\n### Step 1: Install Dependencies and Set Up Environment\n\nInstall Optuna or Ray Tune with appropriate backends.\n\n```bash\n# Create virtual environment\npython -m venv venv\nsource venv\u002Fbin\u002Factivate  # On Windows: venv\\Scripts\\activate\n\n# Option 1: Optuna (simpler, good for single-machine)\npip install optuna optuna-dashboard\npip install scikit-learn xgboost lightgbm\n\n# Option 2: Ray Tune (distributed, good for multi-machine\u002FGPU)\npip install \"ray[tune]\" optuna hyperopt bayesian-optimization\npip install torch torchvision  # if optimizing neural networks\n\n# Visualization and tracking\npip install mlflow tensorboard plotly\n```\n\nCreate project structure:\n\n```bash\nmkdir -p automl\u002F{configs,experiments,models,results}\n```\n\n**Got:** Clean environment with required packages installed, no dependency conflicts.\n\n**If fail:** Use Python 3.8-3.11 (compatibility issues with 3.12+); on CUDA errors install CPU-only versions first; on M1\u002FM2 Mac use conda instead of pip for scikit-learn.\n\n### Step 2: Define Search Space and Objective (Optuna)\n\nCreate configuration for hyperparameter search.\n\n```python\n# automl\u002Foptuna_config.py\nimport optuna\nfrom optuna.pruners import HyperbandPruner\nfrom optuna.samplers import TPESampler\nimport xgboost as xgb\nfrom sklearn.metrics import roc_auc_score, mean_squared_error\nimport numpy as np\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Search space covers reasonable hyperparameter ranges, objective function runs without errors, pruning stops unpromising trials early.\n\n**If fail:** With trials crashing, reduce search space (e.g., lower max n_estimators), verify data has no NaN\u002Finf values, check memory usage (reduce batch size if OOM), ensure eval_metric matches task type.\n\n### Step 3: Run Optimization with Advanced Samplers\n\nExecute hyperparameter search with efficient sampling strategies.\n\n```python\n# automl\u002Frun_optimization.py\nimport optuna\nfrom optuna.samplers import TPESampler, CmaEsSampler, NSGAIISampler\nfrom optuna.pruners import HyperbandPruner, MedianPruner, SuccessiveHalvingPruner\nimport joblib\nimport pandas as pd\nfrom pathlib import Path\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Optimization completes with 50-70% of trials pruned early, best parameters found, visualization plots show convergence.\n\n**If fail:** With no pruning happening, verify objective reports intermediate values correctly; if optimization does not improve, try a different sampler (TPE → CmaES); if it crashes with n_jobs>1, use n_jobs=1 for debugging.\n\n### Step 4: Set Up Ray Tune for Distributed Optimization (Alternative)\n\nUse Ray Tune for multi-GPU or multi-node optimization.\n\n```python\n# automl\u002Fray_tune_config.py\nfrom ray import tune\nfrom ray.tune.schedulers import ASHAScheduler, PopulationBasedTraining\nfrom ray.tune.search.optuna import OptunaSearch\nfrom ray.tune.search import ConcurrencyLimiter\nimport xgboost as xgb\nfrom sklearn.metrics import roc_auc_score\nimport os\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Ray Tune runs trials in parallel across CPUs\u002FGPUs, ASHA scheduler stops bad trials early, best configuration found and logged.\n\n**If fail:** With Ray crashes, start with `ray.init(num_cpus=2, num_gpus=0)` for debugging, reduce concurrent trials if OOM, check that train function does not modify shared data, use `tune.report()` not `return` for metrics.\n\n### Step 5: Track Experiments with MLflow\n\nIntegrate with MLflow for experiment tracking and model registry.\n\n```python\n# automl\u002Fmlflow_tracking.py\nimport mlflow\nimport mlflow.xgboost\nfrom mlflow.tracking import MlflowClient\nimport optuna\nfrom pathlib import Path\n\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** All trials logged to MLflow with parameters and metrics, best model registered in MLflow registry, experiments viewable in MLflow UI.\n\n**If fail:** Start MLflow UI with `mlflow ui --backend-store-uri file:.\u002Fautoml\u002Fmlruns`, check write permissions to mlruns directory, if registration fails verify model registry is configured, ensure model artifact size \u003C 2GB.\n\n### Step 6: Deploy Best Model and Monitor Performance\n\nSave optimized model and set up monitoring.\n\n```python\n# automl\u002Fdeploy_model.py\nimport joblib\nimport json\nfrom pathlib import Path\nimport optuna\nimport xgboost as xgb\n\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Model saved in production-ready format, configuration documented, inference script created for deployment.\n\n**If fail:** With model file too large (>100MB), consider model compression or feature selection, verify model loads correctly in fresh Python session, test inference script with sample data before deployment.\n\n## Validation\n\n- [ ] Optuna\u002FRay Tune installs without dependency conflicts\n- [ ] Search space includes reasonable hyperparameter ranges\n- [ ] Objective function runs successfully for single trial\n- [ ] Optimization completes 50+ trials within time budget\n- [ ] Pruning stops 40-70% of unpromising trials early\n- [ ] Best parameters improve over default configuration by >5%\n- [ ] Visualizations show convergence (optimization history flattens)\n- [ ] MLflow logs all trials with parameters and metrics\n- [ ] Final model saved and loads correctly\n- [ ] Deployment package includes all necessary files\n\n## Pitfalls\n\n- **Overfitting to validation set**: Running 1000s of trials implicitly optimizes for validation set; use holdout test set or time-based split for final evaluation\n- **Ignoring feature engineering**: AutoML finds best hyperparameters but does not create features; invest in feature engineering first\n- **Search space too wide**: Unbounded or very wide ranges waste trials on unrealistic values; use domain knowledge to constrain\n- **Not using early stopping**: Training full epochs for every trial is wasteful; enable early stopping in objective function\n- **Ignoring compute costs**: 100 trials × 10 minutes = 16 hours; consider compute budget when setting n_trials\n- **Categorical features not encoded**: Most algorithms need numeric features; encode categoricals before optimization\n- **Imbalanced data**: Default metrics may mislead with class imbalance; use F1, AUC, or custom metrics\n- **Not saving intermediate results**: Crashes lose all progress; use persistent storage (Optuna SQLite, MLflow) to resume\n\n## Related Skills\n\n- `track-ml-experiments` - MLflow experiment tracking and versioning\n- `orchestrate-ml-pipeline` - Airflow\u002FKubeflow for production AutoML pipelines\n","# Setup AutoML Pipeline\n\n\n> See [Extended Examples](references\u002FEXAMPLES.md) for complete configuration files and templates.\n\nAutomate hyperparameter tuning and model selection using Optuna or Ray Tune with efficient search strategies.\n\n## When to Use\n\n- Starting a new ML project and needing good model configurations quickly\n- Retraining an existing model with new data and re-optimizing hyperparameters\n- Comparing multiple algorithms and their optimal configurations\n- Limited time for manual tuning but need near-optimal performance\n- Team lacks deep expertise in specific algorithm hyperparameters\n- Need a reproducible and documented optimization process\n\n## Inputs\n\n- **Required**: Training dataset with features and labels\n- **Required**: Validation dataset for objective evaluation\n- **Required**: Model type(s) to optimize (e.g., XGBoost, LightGBM, neural network)\n- **Required**: Optimization objective (metric to maximize\u002Fminimize)\n- **Required**: Compute budget (time or number of trials)\n- **Optional**: Search space constraints (min\u002Fmax values for hyperparameters)\n- **Optional**: Prior knowledge of good hyperparameter ranges\n\n## Procedure\n\n### Step 1: Install Dependencies and Set Up Environment\n\nInstall Optuna or Ray Tune with appropriate backends.\n\n```bash\n# Create virtual environment\npython -m venv venv\nsource venv\u002Fbin\u002Factivate  # On Windows: venv\\Scripts\\activate\n\n# Option 1: Optuna (simpler, good for single-machine)\npip install optuna optuna-dashboard\npip install scikit-learn xgboost lightgbm\n\n# Option 2: Ray Tune (distributed, good for multi-machine\u002FGPU)\npip install \"ray[tune]\" optuna hyperopt bayesian-optimization\npip install torch torchvision  # if optimizing neural networks\n\n# Visualization and tracking\npip install mlflow tensorboard plotly\n```\n\nCreate project structure:\n\n```bash\nmkdir -p automl\u002F{configs,experiments,models,results}\n```\n\n**Got:** Clean environment with required packages installed, no dependency conflicts.\n\n**If fail:** Use Python 3.8-3.11 (compatibility issues with 3.12+); on CUDA errors install CPU-only versions first; on M1\u002FM2 Mac use conda instead of pip for scikit-learn.\n\n### Step 2: Define Search Space and Objective (Optuna)\n\nCreate configuration for hyperparameter search.\n\n```python\n# automl\u002Foptuna_config.py\nimport optuna\nfrom optuna.pruners import HyperbandPruner\nfrom optuna.samplers import TPESampler\nimport xgboost as xgb\nfrom sklearn.metrics import roc_auc_score, mean_squared_error\nimport numpy as np\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Search space covers reasonable hyperparameter ranges, objective function runs without errors, pruning stops unpromising trials early.\n\n**If fail:** With trials crashing, reduce search space (e.g., lower max n_estimators), verify data has no NaN\u002Finf values, check memory usage (reduce batch size if OOM), ensure eval_metric matches task type.\n\n### Step 3: Run Optimization with Advanced Samplers\n\nExecute hyperparameter search with efficient sampling strategies.\n\n```python\n# automl\u002Frun_optimization.py\nimport optuna\nfrom optuna.samplers import TPESampler, CmaEsSampler, NSGAIISampler\nfrom optuna.pruners import HyperbandPruner, MedianPruner, SuccessiveHalvingPruner\nimport joblib\nimport pandas as pd\nfrom pathlib import Path\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Optimization completes with 50-70% of trials pruned early, best parameters found, visualization plots show convergence.\n\n**If fail:** With no pruning happening, verify objective reports intermediate values correctly; if optimization does not improve, try a different sampler (TPE → CmaES); if it crashes with n_jobs>1, use n_jobs=1 for debugging.\n\n### Step 4: Set Up Ray Tune for Distributed Optimization (Alternative)\n\nUse Ray Tune for multi-GPU or multi-node optimization.\n\n```python\n# automl\u002Fray_tune_config.py\nfrom ray import tune\nfrom ray.tune.schedulers import ASHAScheduler, PopulationBasedTraining\nfrom ray.tune.search.optuna import OptunaSearch\nfrom ray.tune.search import ConcurrencyLimiter\nimport xgboost as xgb\nfrom sklearn.metrics import roc_auc_score\nimport os\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Ray Tune runs trials in parallel across CPUs\u002FGPUs, ASHA scheduler stops bad trials early, best configuration found and logged.\n\n**If fail:** With Ray crashes, start with `ray.init(num_cpus=2, num_gpus=0)` for debugging, reduce concurrent trials if OOM, check that train function does not modify shared data, use `tune.report()` not `return` for metrics.\n\n### Step 5: Track Experiments with MLflow\n\nIntegrate with MLflow for experiment tracking and model registry.\n\n```python\n# automl\u002Fmlflow_tracking.py\nimport mlflow\nimport mlflow.xgboost\nfrom mlflow.tracking import MlflowClient\nimport optuna\nfrom pathlib import Path\n\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** All trials logged to MLflow with parameters and metrics, best model registered in MLflow registry, experiments viewable in MLflow UI.\n\n**If fail:** Start MLflow UI with `mlflow ui --backend-store-uri file:.\u002Fautoml\u002Fmlruns`, check write permissions to mlruns directory, if registration fails verify model registry is configured, ensure model artifact size \u003C 2GB.\n\n### Step 6: Deploy Best Model and Monitor Performance\n\nSave optimized model and set up monitoring.\n\n```python\n# automl\u002Fdeploy_model.py\nimport joblib\nimport json\nfrom pathlib import Path\nimport optuna\nimport xgboost as xgb\n\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Model saved in production-ready format, configuration documented, inference script created for deployment.\n\n**If fail:** With model file too large (>100MB), consider model compression or feature selection, verify model loads correctly in fresh Python session, test inference script with sample data before deployment.\n\n## Validation\n\n- [ ] Optuna\u002FRay Tune installs without dependency conflicts\n- [ ] Search space includes reasonable hyperparameter ranges\n- [ ] Objective function runs successfully for single trial\n- [ ] Optimization completes 50+ trials within time budget\n- [ ] Pruning stops 40-70% of unpromising trials early\n- [ ] Best parameters improve over default configuration by >5%\n- [ ] Visualizations show convergence (optimization history flattens)\n- [ ] MLflow logs all trials with parameters and metrics\n- [ ] Final model saved and loads correctly\n- [ ] Deployment package includes all necessary files\n\n## Pitfalls\n\n- **Overfitting to validation set**: Running 1000s of trials implicitly optimizes for validation set; use holdout test set or time-based split for final evaluation\n- **Ignoring feature engineering**: AutoML finds best hyperparameters but does not create features; invest in feature engineering first\n- **Search space too wide**: Unbounded or very wide ranges waste trials on unrealistic values; use domain knowledge to constrain\n- **Not using early stopping**: Training full epochs for every trial is wasteful; enable early stopping in objective function\n- **Ignoring compute costs**: 100 trials × 10 minutes = 16 hours; consider compute budget when setting n_trials\n- **Categorical features not encoded**: Most algorithms need numeric features; encode categoricals before optimization\n- **Imbalanced data**: Default metrics may mislead with class imbalance; use F1, AUC, or custom metrics\n- **Not saving intermediate results**: Crashes lose all progress; use persistent storage (Optuna SQLite, MLflow) to resume\n\n## Related Skills\n\n- `track-ml-experiments` - MLflow experiment tracking and versioning\n- `orchestrate-ml-pipeline` - Airflow\u002FKubeflow for production AutoML pipelines\n",{"name":5,"locale":23,"license":14,"metadata":24,"translator":31,"description":32,"allowed-tools":33,"source_commit":34,"source_locale":35,"translation_date":36},"caveman-lite",{"tags":25,"author":26,"domain":27,"version":28,"language":29,"complexity":30},"automl, optuna, ray-tune, hyperparameter, optimization, hyperband, asha","Philipp Thoss","mlops","1.0","multi","intermediate","Julius Brussee homage — caveman","Configure automated ML pipelines using Optuna or Ray Tune for hyperparameter optimization. Implement efficient search strategies (Hyperband, ASHA), define search spaces, and set up early stopping to find optimal model configurations with minimal manual tuning. Use when starting a new ML project, retraining with new data and re-optimizing hyperparameters, comparing multiple algorithms, or when the team lacks deep expertise in specific algorithm hyperparameters.\n","Read Write Edit Bash Grep Glob","82c77053","en","2026-05-03","master","f483e54f3ad5f202a77d45f2e57f8d970b71195f","skills_library\u002Fsetup-automl-pipeline",2,"https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills","https:\u002F\u002Fgithub.com\u002Froohe","yoou.link","2026-05-19 15:56:07",{"source":8,"skillName":5,"commands":46,"prompt":56,"zipAvailable":57},[47,50,53],{"tool":48,"command":49},"npx","npx skills add roohe\u002Fagentic-super-skills --skill setup-automl-pipeline",{"tool":51,"command":52},"bunx","bunx skills add roohe\u002Fagentic-super-skills --skill setup-automl-pipeline",{"tool":54,"command":55},"pnpm","pnpm dlx skills add roohe\u002Fagentic-super-skills --skill setup-automl-pipeline","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fsetup-automl-pipeline\u002FSKILL.md",true,[59],{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002FEXAMPLES.md","REFERENCE","参考文档",26051,"https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fsetup-automl-pipeline\u002Freferences\u002FEXAMPLES.md",[66,73,80,87,94,101],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M73D865CZME2FCMKA4PQM348","zarr-python","Chunked N-D arrays for cloud storage. Compressed arrays, parallel I\u002FO, S3\u002FGCS integration, NumPy\u002FDask\u002FXarray compatible, for large-scale scientific computing pipelines.","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fzarr-python\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":18},"01M73D7Z64J4ABWZJD7P44XBBF","youtube-downloader","Download and process YouTube content for research. Use when: downloading competitor videos for analysis; extracting audio for podcasts; getting transcripts for content repurposing; archiving webinars; research content curation","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fyoutube-downloader\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":18},"01M73D7K8778TGZV3GJFR3703D","yc-pitch-deck","Create a compelling startup pitch deck that follows the structure proven to raise billions from top investors. Master the YC and Sequoia formats that get founders funded. Use when: **Fundraising** to create investor pitch decks; **YC application** to structure your narrative; **Demo Day prep** to craft your 2-minute pitch; **Angel investors** to communicate your opportunity; **Partnership pitches** to structure compelling asks","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fyc-pitch-deck\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":18},"01M73D70KJSE0E8Z28YMNNRT04","x-twitter-scraper","Use when the user needs X (Twitter) data or confirmation-gated X actions through Xquik: tweet search, user lookup, follower extraction, media download, monitoring, webhooks, MCP, SDKs, posting, likes, DMs, and profile updates. Requires a Xquik API key. Never ask for X login material.","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fx-twitter-scraper\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":18},"01M73D6ED01GV077R7K4NZNFZJ","write-vignette","Create R package vignettes using R Markdown or Quarto. Covers vignette setup, YAML configuration, code chunk options, building and testing, and CRAN requirements for vignettes. Use when adding a Getting Started tutorial, documenting complex workflows spanning multiple functions, creating domain-specific guides, or when CRAN submission requires user-facing documentation beyond function help pages.","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fwrite-vignette\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":18},"01M73D69YJR0BK3NFHE0CYG3EF","write-validation-documentation","Write IQ\u002FOQ\u002FPQ validation documentation for computerized systems in regulated environments. Covers protocols, reports, test scripts, deviation handling, and approval workflows. Use when validating R or other software for regulated use, preparing for a regulatory audit, documenting qualification of computing environments, or creating and updating validation protocols and reports for new or re-qualified systems.","https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Fwrite-validation-documentation\u002FSKILL.md",[],[]]