The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes several bash scripts (00-assess.sh, scaffold-dev-tracking.sh, verify.sh) designed to automate the MLOps lifecycle. These scripts perform read-only audits of local codebases, generate infrastructure configuration files, and use the MLflow CLI for environment verification.\n- [EXTERNAL_DOWNLOADS]: The scaffold-dev-tracking.sh script creates a docker-compose.yml file that facilitates pulling official images for PostgreSQL and Python from Docker Hub. It also includes instructions for the container to install the mlflow package directly from PyPI.\n- [REMOTE_CODE_EXECUTION]: The verify.sh script performs health checks by executing short Python commands (python3 -c) that utilize the mlflow library to query the tracking server's status and registry state.\n- [CREDENTIALS_UNSAFE]: The development scaffolding script (scaffold-dev-tracking.sh) defines a default password (mlflow-dev-only) for a local-only PostgreSQL instance. This is a common pattern for local development setups, and the skill documentation explicitly advises users to rotate these credentials and implement proper authentication for production deployments.

mlflow-mlops-migration