Harness-engineering harness-ml-ops

name: harness-ml-ops version: "1.0.0" description: Model serving patterns, experiment tracking, prompt evaluation, and ML pipeline management stability: static cognitive_mode: advisory-guide paths:

• '*.ipynb'
• 'mlflow.yaml' triggers:
• manual
• on_new_feature platforms:
• claude-code
• gemini-cli
• cursor
• codex tools:
• Bash
• Read
• Write
• Edit
• Glob
• Grep
• emit_interaction cli: command: harness skill run harness-ml-ops args:
  • name: path description: Project root path required: false
  • name: focus description: "Audit focus: serving, tracking, evaluation, pipeline, all. Defaults to all." required: false
  • name: framework description: "ML framework: mlflow, wandb, sagemaker, vertex-ai. Auto-detected when omitted." required: false mcp: tool: run_skill input: skill: harness-ml-ops path: string type: rigid tier: 3 internal: false keywords:
• MLOps
• model serving
• experiment tracking
• MLflow
• Weights and Biases
• model registry
• feature store
• prompt evaluation
• LLM
• model deployment
• inference
• training pipeline stack_signals:
• "models/"
• "experiments/"
• "mlflow/"
• "wandb/"
• "src//ml/"
• "src//models/"
• "notebooks/"
• "*.ipynb"
• "prompts/"
• "evals/" phases:
• name: detect description: Identify ML frameworks, model artifacts, experiment configs, and serving infrastructure required: true
• name: analyze description: Evaluate experiment tracking, model versioning, reproducibility, and serving patterns required: true
• name: design description: Recommend pipeline improvements, evaluation frameworks, and deployment strategies required: true
• name: validate description: Verify reproducibility, model registry hygiene, and evaluation coverage required: true state: persistent: false files: [] depends_on: []