Mlops monitoring drift
Skill planifest/planifest-framework/planifest-framework/external-skills/mlops-monitoring-drift
A specification framework for agentic development. Agents build from complete specs - not guesses.
npx -y skills add planifest/planifest-framework --skill mlops-monitoring-driftAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
MLOps drift monitoring workflow for detecting data drift, concept drift, and quality degradation with actionable response rules. Use when production ML systems need drift detection thresholds and escalation ownership; do not use for model-architecture research decisions.
SKILL.md
1.8 KB, 324 tokens by cl100k_base, as published. Nobody here has run it
Mlops Monitoring Drift
Overview
Use this skill to detect meaningful model degradation early and trigger appropriate remediation actions.
Scope Boundaries
- Use this skill when the task matches the trigger condition described in
description. - Do not use this skill when the primary task falls outside this skill's domain.
Shared References
- Drift alerting and escalation rules:
references/drift-alerting-escalation-rules.md
Templates And Assets
- Drift monitoring policy template:
assets/drift-monitoring-policy-template.md
Inputs To Gather
- Drift signals and quality metrics to monitor.
- Alert thresholds and acceptable noise level.
- Escalation owners and response SLA.
- Retraining and rollback policies.
Deliverables
- Drift monitoring policy and thresholds.
- Alert routing and severity model.
- Response playbook for drift events.
Workflow
- Define monitoring policy in
assets/drift-monitoring-policy-template.md. - Validate threshold actionability via
references/drift-alerting-escalation-rules.md. - Test alert behavior with historical replay or backtests.
- Assign response ownership and SLA per severity.
- Publish retraining/mitigation decision criteria.
Quality Standard
- Alerts are actionable, not noise-heavy.
- Severity levels map to clear response ownership.
- Retraining triggers are explicit and auditable.
Failure Conditions
- Stop when drift thresholds are not operationally actionable.
- Stop when alerts have no clear owner.
- Escalate when degradation risk remains unmanaged.