aeon

This skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring specialized algorithms beyond standard ML approaches. Particularly suited for univariate and multivariate time series analysis with scikit-learn compatible APIs.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with executable code for every capability, but it is held back by content duplication, the absence of explicit validation checkpoints in its workflows, and — most seriously — eleven referenced reference files that do not exist in the bundle. Progressive disclosure is structurally sound in intent but unfulfilled because the bundle is missing.

Suggestions

Remove duplicated content: consolidate the second STOMP example and the second RocketTransformer example, and merge the inline 'Algorithm Selection' list with the later 'Algorithm Selection Guide' so each topic appears once.

Add explicit validation checkpoints to the Common Workflows (e.g., assert data shape is (n_samples, n_channels, n_timepoints) before fit; check model score / residual diagnostics before relying on predictions) to turn linear pipelines into feedback-looped workflows.

Provide the missing references/ bundle files (classification.md, regression.md, clustering.md, forecasting.md, anomaly_detection.md, segmentation.md, similarity_search.md, transformations.md, distances.md, networks.md, datasets_benchmarking.md) or remove the dangling references so signaled navigation resolves to real content.

Dimension	Reasoning	Score
Conciseness	Mostly code-focused and assumes Claude's competence, but includes real duplication — STOMP anomaly detection appears in section 5 and again in 'Anomaly Detection with Visualization', RocketTransformer appears in 'Feature Extraction' and again in 'Common Workflows', and algorithm-selection guidance is repeated inline and in the 'Algorithm Selection Guide' — so it could be tightened; not score 3 because not every token earns its place.	2 / 3
Actionability	Fully executable, copy-paste-ready code with real imports and class names across all seven capabilities (e.g., RocketClassifier, TimeSeriesKMeans, ARIMA, STOMP), plus concrete selection tables and the data-shape contract '(n_samples, n_channels, n_timepoints)'; not score 2 since examples are complete rather than pseudocode.	3 / 3
Workflow Clarity	Common-workflow pipelines (classification pipeline, feature-extraction + traditional ML) are sequenced, and Best Practices mentions validation/compare-baselines as advice, but there are no explicit validation checkpoints or error-recovery feedback loops in the workflows; not score 1 because sequences are present, not score 3 because checkpoints are missing/implicit.	2 / 3
Progressive Disclosure	The body is well organized and signals one-level-deep references (classification.md, regression.md, etc. plus a Reference Documentation list), but the referenced bundle files do not exist — the references/ directory is absent — so the navigation is broken and the score-3 'easy navigation' requirement fails; not score 1 because signaling and one-level depth are good in intent.	2 / 3
	Total	9 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it enumerates seven concrete capabilities, provides an explicit 'Use when...' trigger covering natural user terms, and carves out a distinct time-series-ML niche. Voice is third-person/imperative ('This skill should be used', 'Use when'), so no person-voice penalty applies.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search' — matching the score-3 anchor of several specific concrete actions rather than vague language.	3 / 3
Completeness	Explicitly answers both what ('time series machine learning tasks including...') and when ('Use when working with temporal data, sequential patterns, or time-indexed observations...'), matching the score-3 what-AND-when-with-explicit-triggers anchor; not score 2 because the 'when' is explicit, not implied.	3 / 3
Trigger Term Quality	Covers natural terms users would say ('time series', 'forecasting', 'anomaly detection', 'clustering', 'temporal data'); not merely score-2 'some relevant keywords' since it spans the full task vocabulary users invoke.	3 / 3
Distinctiveness Conflict Risk	Clear niche — time series ML 'requiring specialized algorithms beyond standard ML approaches' with scikit-learn-compatible APIs — unlikely to trigger for generic ML skills; the specialization prevents the overlap that would cap it at score 2.	3 / 3
	Total	12 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
referenced_paths_exist	Referenced path issues: 11 missing	Warning

	Total	14 / 16 Passed

Repository: wu-yc/LabClaw
Path: skills/general/aeon/SKILL.md
Commit: df37802

Reviewed: about 2 hours ago

Table of Contents

Discovery Implementation Validation

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.