Ml Pipeline Starter

Build and deploy production ML pipelines with data processing, model training, evaluation, and deployment using TensorFlow, PyTorch, or Scikit-learn.

Security Scan

The skill's description fits an ML pipeline helper, but its runtime instructions tell the user/agent to execute a local script (./ml-pipeline.sh) that is not included or sourced from a trusted location — this gap is suspicious and could lead to executing unknown code.

Purpose & Capability

The stated purpose (build/deploy ML pipelines) matches the features listed and required tools (Python, Docker). However, the SKILL.md instructs running a local script (./ml-pipeline.sh) that is not provided, linked, or explained; an instruction-only skill that requires an external script without direction is incoherent and raises questions about how the capability is delivered.

Instruction Scope

The runtime instructions directly call ./ml-pipeline.sh (create/train/deploy). Because the skill bundle contains no script, the instructions implicitly rely on a script on the user's system or elsewhere. That gives the agent/user a broad implicit permission to execute arbitrary shell operations via that script; SKILL.md provides no details of the script's behavior, no safety checks, and no source to verify.

Install Mechanism

No install spec is present (instruction-only), so nothing will be written to disk by the skill itself. This is low-risk in terms of automatic installation, but combined with missing script content it shifts risk to whatever ./ml-pipeline.sh the user runs.

Credentials

The skill requests no environment variables, credentials, or config paths. That is proportionate to the described functionality and is a positive signal.

Persistence & Privilege

The skill is not always-included and uses normal user-invocable/autonomous settings. It does not request elevated persistence or modify other skills/configurations.

What to consider before installing

This skill is suspicious because it tells you to run ./ml-pipeline.sh but does not include that script or state where it comes from. Before installing or following its instructions: (1) ask the author for the actual ml-pipeline.sh source or a trusted repository link and review its contents; (2) never run unknown shell scripts on your host — inspect them and run them in an isolated environment (container or VM) first; (3) prefer skills that include their tooling or point to official releases (GitHub, PyPI, Docker Hub) and provide installation steps you can audit; (4) if you must run a script from an external source, verify its integrity (checksum/signature) and review for network/exfiltration or privileged system operations. If the author cannot provide the script or a verifiable source, treat the skill as unsafe to run.

Build production ML pipelines.

Features

Data Processing

• Data validation
• Feature engineering
• Data augmentation

Model Training

• Hyperparameter tuning
• Cross-validation
• Model versioning

Evaluation

• Metrics tracking
• Bias detection
• Performance monitoring

Deployment

• Model serving
• A/B testing
• Rollback support

Quick Start

# Create pipeline
./ml-pipeline.sh create my-model

# Train
./ml-pipeline.sh train my-model

# Deploy
./ml-pipeline.sh deploy my-model production

Frameworks

• TensorFlow
• PyTorch
• Scikit-learn

Requirements

• Python 3.8+
• Docker

Author

Sunshine-del-ux