model-drift-detector
Model Drift Detector - Auto-activating skill for ML Deployment. Triggers on: model drift detector, model drift detector Part of the ML Deployment skill category.
36
3%
Does it follow best practices?
Impact
99%
1.02xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./planned-skills/generated/08-ml-deployment/model-drift-detector/SKILL.mdQuality
Discovery
7%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is extremely weak across all dimensions. It is essentially a title and category label with no concrete actions, no meaningful trigger terms, and no explicit guidance on when Claude should select this skill. The repeated trigger term adds no value.
Suggestions
Add specific concrete actions the skill performs, e.g., 'Detects statistical drift in model predictions, compares feature distributions between training and production data, generates drift severity reports, and recommends retraining schedules.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user mentions model drift, data drift, concept drift, prediction degradation, model monitoring, feature distribution shifts, or production model performance decline.'
Remove the duplicated trigger term and expand with diverse keyword variations users would naturally use, such as 'model staleness', 'distribution shift', 'covariate shift', '.pkl model monitoring'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names a domain ('ML Deployment') and a concept ('Model Drift Detector') but does not describe any concrete actions. There are no specific capabilities listed such as 'monitors prediction distributions', 'compares training vs production data', or 'generates drift reports'. | 1 / 3 |
Completeness | The description fails to answer 'what does this do' beyond naming itself, and the 'when' clause is essentially just the skill name repeated. There is no explicit 'Use when...' guidance with meaningful triggers. | 1 / 3 |
Trigger Term Quality | The trigger terms are just 'model drift detector' repeated twice. There are no natural keyword variations a user might say, such as 'data drift', 'concept drift', 'model monitoring', 'prediction shift', 'feature drift', or 'model degradation'. | 1 / 3 |
Distinctiveness Conflict Risk | The term 'model drift detector' is somewhat specific to a niche domain, which reduces conflict risk with unrelated skills. However, within an ML Deployment category, it's unclear how this differs from other monitoring or deployment skills due to the lack of concrete capability descriptions. | 2 / 3 |
Total | 5 / 12 Passed |
Implementation
0%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is essentially a placeholder with no substantive content. It contains no actionable guidance, no code examples, no specific techniques for detecting model drift (e.g., PSI, KS test, KL divergence), and no workflow for implementing drift detection in production. Every section restates the same vague description without teaching Claude anything it doesn't already know.
Suggestions
Add concrete, executable code examples for drift detection techniques (e.g., Population Stability Index, Kolmogorov-Smirnov test, or using libraries like evidently/alibi-detect)
Define a clear multi-step workflow: collect baseline distributions → monitor incoming data → compute drift metrics → alert/retrain thresholds → validation steps
Replace the meta-description sections (Purpose, When to Use, Capabilities, Example Triggers) with actual technical content including specific drift metrics, threshold recommendations, and production monitoring patterns
Add concrete configuration examples for drift monitoring tools (e.g., Evidently dashboards, custom Prometheus metrics, or MLflow model registry integration)
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is entirely filler and meta-description. It explains what the skill does in abstract terms without providing any actual technical content. Every section restates the same vague information about 'model drift detector' without adding substance. | 1 / 3 |
Actionability | There is zero concrete guidance—no code, no commands, no specific algorithms, no statistical tests, no thresholds, no library recommendations. The skill describes rather than instructs, offering only vague promises like 'provides step-by-step guidance' without actually providing any. | 1 / 3 |
Workflow Clarity | No workflow is defined at all. There are no steps, no sequence, no validation checkpoints. The content merely states it can provide guidance without actually laying out any process for detecting model drift. | 1 / 3 |
Progressive Disclosure | The content is a flat, repetitive document with no meaningful structure. Sections like 'Capabilities' and 'Example Triggers' repeat the same information. There are no references to detailed materials, examples, or related documentation. | 1 / 3 |
Total | 4 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
87f14eb
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.