[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M73AS34WYFX7YN12C4E1ECQ6":3,"skill-same:01M73AS34WYFX7YN12C4E1ECQ6":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},"01M73AS34WYFX7YN12C4E1ECQ6","orchestrate-ml-pipeline",null,"Orchestrate end-to-end machine learning pipelines using Prefect or Airflow with DAG construction, task dependencies, retry logic, scheduling, monitoring, and integration with MLflow, DVC, and feature stores for production ML workflows. Use when automating multi-step ML workflows from data ingestion to deployment, scheduling periodic model retraining, coordinating distributed training tasks, or managing retry logic and failure recovery across pipeline stages.","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\u002Forchestrate-ml-pipeline\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: orchestrate-ml-pipeline\nlocale: caveman-lite\nsource_locale: en\nsource_commit: 82c77053\ntranslator: \"Julius Brussee homage — caveman\"\ntranslation_date: \"2026-04-26\"\ndescription: >\n  Orchestrate end-to-end machine learning pipelines using Prefect or Airflow with\n  DAG construction, task dependencies, retry logic, scheduling, monitoring, and\n  integration with MLflow, DVC, and feature stores for production ML workflows. Use\n  when automating multi-step ML workflows from data ingestion to deployment, scheduling\n  periodic model retraining, coordinating distributed training tasks, or managing retry\n  logic and failure recovery across pipeline stages.\nlicense: MIT\nallowed-tools: Read Write Edit Bash Grep Glob\nmetadata:\n  author: Philipp Thoss\n  version: \"1.0\"\n  domain: mlops\n  complexity: advanced\n  language: multi\n  tags: prefect, airflow, pipeline, dag, orchestration\n---\n\n# Orchestrate ML Pipeline\n\n\n> See [Extended Examples](references\u002FEXAMPLES.md) for complete configuration files and templates.\n\nBuild and orchestrate end-to-end machine learning pipelines with dependency management, scheduling, and monitoring.\n\n## When to Use\n\n- Automating multi-step ML workflows from data ingestion to deployment\n- Scheduling periodic model retraining on fresh data\n- Coordinating distributed data processing and training tasks\n- Implementing complex dependencies between ML pipeline stages\n- Managing retry logic and failure recovery\n- Monitoring pipeline execution and alerting on failures\n- Orchestrating feature engineering, training, evaluation, and deployment\n- Building reproducible ML workflows across environments\n\n## Inputs\n\n- **Required**: ML pipeline components (data ingestion, preprocessing, training, evaluation)\n- **Required**: Orchestration framework choice (Prefect, Airflow, Kubeflow)\n- **Required**: Python environment with orchestration library installed\n- **Optional**: Kubernetes cluster for distributed execution\n- **Optional**: MLflow tracking server for experiment logging\n- **Optional**: DVC for data versioning\n- **Optional**: Slack\u002Femail for alerting\n- **Optional**: Monitoring infrastructure (Prometheus, Grafana)\n\n## Procedure\n\n### Step 1: Choose and Install Orchestration Framework\n\nSelect appropriate framework and set up infrastructure.\n\n```bash\n# Option 1: Prefect (modern, Pythonic, simpler)\npip install prefect\npip install prefect-aws prefect-dask prefect-docker\n\n# Start Prefect server (local development)\nprefect server start\n\n# Or use Prefect Cloud (managed)\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nDocker Compose for Airflow:\n\n```yaml\n# docker-compose.airflow.yml\nversion: '3.8'\n\nx-airflow-common: &airflow-common\n  image: apache\u002Fairflow:2.8.0\n  environment:\n    AIRFLOW__CORE__EXECUTOR: CeleryExecutor\n    AIRFLOW__DATABASE__SQL_ALCHEMY_CONN: postgresql+psycopg2:\u002F\u002Fairflow:airflow@postgres\u002Fairflow\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Orchestration framework installed, web UI accessible (Prefect at http:\u002F\u002Flocalhost:4200, Airflow at http:\u002F\u002Flocalhost:8080), database initialized, scheduler running.\n\n**If fail:** Check port availability (`netstat -tulpn | grep 8080`), verify database connection, ensure Redis running for Celery, check Python version compatibility (Airflow requires ≥3.8), verify Docker daemon for containerized setup, inspect logs for initialization errors.\n\n### Step 2: Build ML Pipeline with Prefect\n\nCreate Prefect flow with tasks for each pipeline stage.\n\n```python\n# prefect_ml_pipeline.py\nfrom prefect import flow, task\nfrom prefect.tasks import task_input_hash\nfrom datetime import timedelta\nimport pandas as pd\nimport mlflow\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.model_selection import train_test_split\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nDeploy and schedule:\n\n```python\n# deploy_prefect.py\nfrom prefect.deployments import Deployment\nfrom prefect.server.schemas.schedules import CronSchedule\nfrom prefect_ml_pipeline import ml_training_pipeline\n\n# Create deployment with schedule\ndeployment = Deployment.build_from_flow(\n    flow=ml_training_pipeline,\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Prefect flow executes all tasks in correct order, task failures trigger retries automatically, successful runs show green in UI, MLflow logs experiments, model registered and deployed.\n\n**If fail:** Check task dependencies defined correctly, verify MLflow server accessible, ensure data source paths correct, check for circular dependencies, verify task timeout limits, inspect Prefect logs for detailed errors, check resource availability (memory\u002FCPU).\n\n### Step 3: Build ML Pipeline with Airflow\n\nCreate Airflow DAG for production ML workflow.\n\n```python\n# dags\u002Fml_training_dag.py\nfrom airflow import DAG\nfrom airflow.operators.python import PythonOperator\nfrom airflow.providers.docker.operators.docker import DockerOperator\nfrom airflow.utils.dates import days_ago\nfrom datetime import datetime, timedelta\nimport mlflow\nimport pandas as pd\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** DAG appears in Airflow UI, scheduled runs execute on time, task failures trigger retries and alerts, XCom passes data between tasks, MLflow integration logs experiments.\n\n**If fail:** Check DAG file syntax (`python dags\u002Fml_training_dag.py`), verify imports available in Airflow environment, ensure XCom not exceeding size limits (use file paths for large data), check email configuration for alerts, verify scheduler running, inspect task logs in Airflow UI.\n\n### Step 4: Implement Advanced Features\n\nAdd dynamic DAGs, branching, and parallel execution.\n\n```python\n# advanced_pipeline.py (Prefect)\nfrom prefect import flow, task\nfrom prefect.task_runners import DaskTaskRunner, ConcurrentTaskRunner\nimport time\n\n@task\ndef process_shard(shard_id: int, data: list) -> dict:\n    \"\"\"Process data shard in parallel.\"\"\"\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nAirflow branching:\n\n```python\n# Airflow branching with BranchPythonOperator\nfrom airflow.operators.python import BranchPythonOperator\n\ndef check_data_quality(**context):\n    \"\"\"Decide which branch to take.\"\"\"\n    data_path = context['ti'].xcom_pull(key='data_path')\n    df = pd.read_csv(data_path)\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Parallel tasks execute concurrently (faster pipeline), conditional branches execute based on logic, dynamic task generation works, Dask cluster distributes work.\n\n**If fail:** Check Dask cluster configured and accessible, verify task_runner specified, ensure branching returns valid task IDs, check for resource contention with parallel tasks, verify conditional logic correctness.\n\n### Step 5: Integrate Monitoring and Alerting\n\nAdd comprehensive monitoring and failure notifications.\n\n```python\n# monitoring_integration.py\nfrom prefect.blocks.notifications import SlackWebhook\nfrom prefect import flow, task, get_run_logger\nfrom prefect.context import FlowRunContext\n\n@task(on_failure=[notify_failure])\ndef critical_task():\n    \"\"\"Task with failure notification.\"\"\"\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nAirflow monitoring with sensors:\n\n```python\n# Airflow SLA and monitoring\nfrom airflow.sensors.base import BaseSensorOperator\nfrom airflow.utils.decorators import apply_defaults\n\ndefault_args = {\n    'sla': timedelta(hours=4),  # Alert if task exceeds 4 hours\n    'on_failure_callback': slack_alert_failure,\n    'on_success_callback': slack_alert_success,\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Slack\u002Femail notifications sent on failures, SLA violations trigger alerts, custom metrics tracked, logs aggregated in monitoring system.\n\n**If fail:** Verify Slack webhook configured correctly, check email SMTP settings, ensure notification blocks loaded properly, verify SLA values reasonable, check for network issues blocking notifications.\n\n### Step 6: Implement CI\u002FCD for Pipelines\n\nVersion control and automate pipeline deployments.\n\n```yaml\n# .github\u002Fworkflows\u002Fdeploy-pipeline.yml\nname: Deploy ML Pipeline\n\non:\n  push:\n    branches: [main]\n    paths:\n      - 'pipelines\u002F**'\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Pipeline tests pass before deployment, automated deployment to production, team notified on successful deployment, pipeline versioning tracked in Git.\n\n**If fail:** Check test coverage and failures, verify Prefect Cloud credentials, ensure deployment script handles errors, check Slack webhook configuration, inspect CI logs for deployment errors.\n\n## Validation\n\n- [ ] Orchestration framework installed and running\n- [ ] Pipeline DAG defined with correct dependencies\n- [ ] All tasks execute in proper order\n- [ ] Retry logic functions correctly on failures\n- [ ] Scheduled runs execute on time\n- [ ] MLflow integration logs experiments\n- [ ] DVC integration versions data\n- [ ] Parallel tasks execute concurrently\n- [ ] Conditional branches work correctly\n- [ ] Monitoring and alerting functional\n- [ ] CI\u002FCD pipeline deploys automatically\n- [ ] Pipeline reproducible across environments\n\n## Pitfalls\n\n- **Circular dependencies**: Task A depends on B, B depends on A — carefully design DAG structure, use Airflow\u002FPrefect validators\n- **Memory leaks**: Long-running tasks accumulate memory — set task timeouts, monitor resource usage, restart workers periodically\n- **XCom size limits**: Passing large data via XCom — use file paths or external storage (S3) instead of direct serialization\n- **Timezone confusion**: Schedule runs at wrong times — always use UTC, explicitly set timezone in schedule\n- **Missing retries**: Tasks fail permanently on transient errors — configure retries with exponential backoff\n- **Tight coupling**: Tasks directly depend on implementation details — use clear interfaces, pass parameters explicitly\n- **No idempotency**: Re-running tasks causes duplicates or errors — design tasks to be idempotent (safe to retry)\n- **Poor error handling**: Failures do not provide useful context — add detailed logging, capture exceptions properly\n- **Resource contention**: Parallel tasks overwhelm resources — limit concurrency, set resource quotas\n- **Version conflicts**: Different tasks need incompatible dependencies — use Docker containers for task isolation\n\n## Related Skills\n\n- `track-ml-experiments` - Integrate MLflow tracking into pipeline tasks\n- `version-ml-data` - Use DVC for data versioning in pipelines\n- `build-feature-store` - Materialize features as pipeline task\n- `deploy-ml-model-serving` - Add deployment as final pipeline stage\n- `deploy-to-kubernetes` - Run orchestrated pipelines on Kubernetes\n","# Orchestrate ML Pipeline\n\n\n> See [Extended Examples](references\u002FEXAMPLES.md) for complete configuration files and templates.\n\nBuild and orchestrate end-to-end machine learning pipelines with dependency management, scheduling, and monitoring.\n\n## When to Use\n\n- Automating multi-step ML workflows from data ingestion to deployment\n- Scheduling periodic model retraining on fresh data\n- Coordinating distributed data processing and training tasks\n- Implementing complex dependencies between ML pipeline stages\n- Managing retry logic and failure recovery\n- Monitoring pipeline execution and alerting on failures\n- Orchestrating feature engineering, training, evaluation, and deployment\n- Building reproducible ML workflows across environments\n\n## Inputs\n\n- **Required**: ML pipeline components (data ingestion, preprocessing, training, evaluation)\n- **Required**: Orchestration framework choice (Prefect, Airflow, Kubeflow)\n- **Required**: Python environment with orchestration library installed\n- **Optional**: Kubernetes cluster for distributed execution\n- **Optional**: MLflow tracking server for experiment logging\n- **Optional**: DVC for data versioning\n- **Optional**: Slack\u002Femail for alerting\n- **Optional**: Monitoring infrastructure (Prometheus, Grafana)\n\n## Procedure\n\n### Step 1: Choose and Install Orchestration Framework\n\nSelect appropriate framework and set up infrastructure.\n\n```bash\n# Option 1: Prefect (modern, Pythonic, simpler)\npip install prefect\npip install prefect-aws prefect-dask prefect-docker\n\n# Start Prefect server (local development)\nprefect server start\n\n# Or use Prefect Cloud (managed)\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nDocker Compose for Airflow:\n\n```yaml\n# docker-compose.airflow.yml\nversion: '3.8'\n\nx-airflow-common: &airflow-common\n  image: apache\u002Fairflow:2.8.0\n  environment:\n    AIRFLOW__CORE__EXECUTOR: CeleryExecutor\n    AIRFLOW__DATABASE__SQL_ALCHEMY_CONN: postgresql+psycopg2:\u002F\u002Fairflow:airflow@postgres\u002Fairflow\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Orchestration framework installed, web UI accessible (Prefect at http:\u002F\u002Flocalhost:4200, Airflow at http:\u002F\u002Flocalhost:8080), database initialized, scheduler running.\n\n**If fail:** Check port availability (`netstat -tulpn | grep 8080`), verify database connection, ensure Redis running for Celery, check Python version compatibility (Airflow requires ≥3.8), verify Docker daemon for containerized setup, inspect logs for initialization errors.\n\n### Step 2: Build ML Pipeline with Prefect\n\nCreate Prefect flow with tasks for each pipeline stage.\n\n```python\n# prefect_ml_pipeline.py\nfrom prefect import flow, task\nfrom prefect.tasks import task_input_hash\nfrom datetime import timedelta\nimport pandas as pd\nimport mlflow\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.model_selection import train_test_split\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nDeploy and schedule:\n\n```python\n# deploy_prefect.py\nfrom prefect.deployments import Deployment\nfrom prefect.server.schemas.schedules import CronSchedule\nfrom prefect_ml_pipeline import ml_training_pipeline\n\n# Create deployment with schedule\ndeployment = Deployment.build_from_flow(\n    flow=ml_training_pipeline,\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Prefect flow executes all tasks in correct order, task failures trigger retries automatically, successful runs show green in UI, MLflow logs experiments, model registered and deployed.\n\n**If fail:** Check task dependencies defined correctly, verify MLflow server accessible, ensure data source paths correct, check for circular dependencies, verify task timeout limits, inspect Prefect logs for detailed errors, check resource availability (memory\u002FCPU).\n\n### Step 3: Build ML Pipeline with Airflow\n\nCreate Airflow DAG for production ML workflow.\n\n```python\n# dags\u002Fml_training_dag.py\nfrom airflow import DAG\nfrom airflow.operators.python import PythonOperator\nfrom airflow.providers.docker.operators.docker import DockerOperator\nfrom airflow.utils.dates import days_ago\nfrom datetime import datetime, timedelta\nimport mlflow\nimport pandas as pd\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** DAG appears in Airflow UI, scheduled runs execute on time, task failures trigger retries and alerts, XCom passes data between tasks, MLflow integration logs experiments.\n\n**If fail:** Check DAG file syntax (`python dags\u002Fml_training_dag.py`), verify imports available in Airflow environment, ensure XCom not exceeding size limits (use file paths for large data), check email configuration for alerts, verify scheduler running, inspect task logs in Airflow UI.\n\n### Step 4: Implement Advanced Features\n\nAdd dynamic DAGs, branching, and parallel execution.\n\n```python\n# advanced_pipeline.py (Prefect)\nfrom prefect import flow, task\nfrom prefect.task_runners import DaskTaskRunner, ConcurrentTaskRunner\nimport time\n\n@task\ndef process_shard(shard_id: int, data: list) -> dict:\n    \"\"\"Process data shard in parallel.\"\"\"\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nAirflow branching:\n\n```python\n# Airflow branching with BranchPythonOperator\nfrom airflow.operators.python import BranchPythonOperator\n\ndef check_data_quality(**context):\n    \"\"\"Decide which branch to take.\"\"\"\n    data_path = context['ti'].xcom_pull(key='data_path')\n    df = pd.read_csv(data_path)\n\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Parallel tasks execute concurrently (faster pipeline), conditional branches execute based on logic, dynamic task generation works, Dask cluster distributes work.\n\n**If fail:** Check Dask cluster configured and accessible, verify task_runner specified, ensure branching returns valid task IDs, check for resource contention with parallel tasks, verify conditional logic correctness.\n\n### Step 5: Integrate Monitoring and Alerting\n\nAdd comprehensive monitoring and failure notifications.\n\n```python\n# monitoring_integration.py\nfrom prefect.blocks.notifications import SlackWebhook\nfrom prefect import flow, task, get_run_logger\nfrom prefect.context import FlowRunContext\n\n@task(on_failure=[notify_failure])\ndef critical_task():\n    \"\"\"Task with failure notification.\"\"\"\n# ... (see EXAMPLES.md for complete implementation)\n```\n\nAirflow monitoring with sensors:\n\n```python\n# Airflow SLA and monitoring\nfrom airflow.sensors.base import BaseSensorOperator\nfrom airflow.utils.decorators import apply_defaults\n\ndefault_args = {\n    'sla': timedelta(hours=4),  # Alert if task exceeds 4 hours\n    'on_failure_callback': slack_alert_failure,\n    'on_success_callback': slack_alert_success,\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Slack\u002Femail notifications sent on failures, SLA violations trigger alerts, custom metrics tracked, logs aggregated in monitoring system.\n\n**If fail:** Verify Slack webhook configured correctly, check email SMTP settings, ensure notification blocks loaded properly, verify SLA values reasonable, check for network issues blocking notifications.\n\n### Step 6: Implement CI\u002FCD for Pipelines\n\nVersion control and automate pipeline deployments.\n\n```yaml\n# .github\u002Fworkflows\u002Fdeploy-pipeline.yml\nname: Deploy ML Pipeline\n\non:\n  push:\n    branches: [main]\n    paths:\n      - 'pipelines\u002F**'\n# ... (see EXAMPLES.md for complete implementation)\n```\n\n**Got:** Pipeline tests pass before deployment, automated deployment to production, team notified on successful deployment, pipeline versioning tracked in Git.\n\n**If fail:** Check test coverage and failures, verify Prefect Cloud credentials, ensure deployment script handles errors, check Slack webhook configuration, inspect CI logs for deployment errors.\n\n## Validation\n\n- [ ] Orchestration framework installed and running\n- [ ] Pipeline DAG defined with correct dependencies\n- [ ] All tasks execute in proper order\n- [ ] Retry logic functions correctly on failures\n- [ ] Scheduled runs execute on time\n- [ ] MLflow integration logs experiments\n- [ ] DVC integration versions data\n- [ ] Parallel tasks execute concurrently\n- [ ] Conditional branches work correctly\n- [ ] Monitoring and alerting functional\n- [ ] CI\u002FCD pipeline deploys automatically\n- [ ] Pipeline reproducible across environments\n\n## Pitfalls\n\n- **Circular dependencies**: Task A depends on B, B depends on A — carefully design DAG structure, use Airflow\u002FPrefect validators\n- **Memory leaks**: Long-running tasks accumulate memory — set task timeouts, monitor resource usage, restart workers periodically\n- **XCom size limits**: Passing large data via XCom — use file paths or external storage (S3) instead of direct serialization\n- **Timezone confusion**: Schedule runs at wrong times — always use UTC, explicitly set timezone in schedule\n- **Missing retries**: Tasks fail permanently on transient errors — configure retries with exponential backoff\n- **Tight coupling**: Tasks directly depend on implementation details — use clear interfaces, pass parameters explicitly\n- **No idempotency**: Re-running tasks causes duplicates or errors — design tasks to be idempotent (safe to retry)\n- **Poor error handling**: Failures do not provide useful context — add detailed logging, capture exceptions properly\n- **Resource contention**: Parallel tasks overwhelm resources — limit concurrency, set resource quotas\n- **Version conflicts**: Different tasks need incompatible dependencies — use Docker containers for task isolation\n\n## Related Skills\n\n- `track-ml-experiments` - Integrate MLflow tracking into pipeline tasks\n- `version-ml-data` - Use DVC for data versioning in pipelines\n- `build-feature-store` - Materialize features as pipeline task\n- `deploy-ml-model-serving` - Add deployment as final pipeline stage\n- `deploy-to-kubernetes` - Run orchestrated pipelines on Kubernetes\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},"prefect, airflow, pipeline, dag, orchestration","Philipp Thoss","mlops","1.0","multi","advanced","Julius Brussee homage — caveman","Orchestrate end-to-end machine learning pipelines using Prefect or Airflow with DAG construction, task dependencies, retry logic, scheduling, monitoring, and integration with MLflow, DVC, and feature stores for production ML workflows. Use when automating multi-step ML workflows from data ingestion to deployment, scheduling periodic model retraining, coordinating distributed training tasks, or managing retry logic and failure recovery across pipeline stages.\n","Read Write Edit Bash Grep Glob","82c77053","en","2026-04-26","master","f483e54f3ad5f202a77d45f2e57f8d970b71195f","skills_library\u002Forchestrate-ml-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 orchestrate-ml-pipeline",{"tool":51,"command":52},"bunx","bunx skills add roohe\u002Fagentic-super-skills --skill orchestrate-ml-pipeline",{"tool":54,"command":55},"pnpm","pnpm dlx skills add roohe\u002Fagentic-super-skills --skill orchestrate-ml-pipeline","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Forchestrate-ml-pipeline\u002FSKILL.md",true,[59],{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002FEXAMPLES.md","REFERENCE","参考文档",19868,"https:\u002F\u002Fgithub.com\u002Froohe\u002Fagentic-super-skills\u002Fblob\u002Fmaster\u002Fskills_library\u002Forchestrate-ml-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",[],[]]