detecting-data-anomalies

Investigate outliers, rare events, spikes, and suspicious records in datasets. Use as an explicit anomaly-analysis helper when you want concrete anomaly-detection workflow guidance, not generic data validation or end-to-end ML ownership.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./bundled/skills/detecting-data-anomalies/SKILL.md

Quality

Content

46%Weight 40%Scale 1-5

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

The SKILL.md body is concise and well-sectioned but largely aspirational: it promises methods, checklists, tables, and plots without executable guidance, and its bundle is a set of identical placeholder scripts plus README pointers to missing files. Navigation between the body and the actual bundle is weak and partly broken.

Suggestions

Add at least one concrete executable example (e.g., a short Isolation Forest snippet with contamination and a threshold) so the skill is actionable rather than descriptive.

Replace the four identical placeholder scripts with real implementations, or remove them and inline minimal working code, since they currently advertise capabilities they do not provide.

Fix progressive disclosure by either creating the referenced files (anomaly_detection_algorithms.md, visualization_templates/, example_datasets/) or removing their README entries, and link the body's bundle (scripts/, references/implementation.md) explicitly from the When-to-use/Outputs sections.

Deliver the promised 'review checklist for false positives and false negatives' and add an explicit validate->fix->retry checkpoint after running detection, which is needed for batch/destructive analysis workflows.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence with no concept over-explanation, but the scripts it advertises are identical auto-generated templates that add no real signal, a minor inefficiency.	4 / 5
Actionability	The body gives only high-level directions ('Apply selected algorithm using Bash tool', 'Generate visualizations') with no executable code, commands, or specific API usage, and the referenced scripts are placeholder templates that do not perform their named functions.	2 / 5
Workflow Clarity	A five-step sequence is listed and a checklist is promised, but there are no explicit validation checkpoints or feedback loops, and the promised 'review checklist for false positives and false negatives' is never delivered.	3 / 5
Progressive Disclosure	Structure exists but references are misaligned: the body cites scikit-learn and creating-data-visualizations rather than its own bundle, while references/README and assets/README list files (anomaly_detection_algorithms.md, visualization_templates/, example_datasets/) that do not exist and are not linked from the body.	3 / 5
	Total	12 / 20 Passed

Description

91%Weight 40%Scale 1-5

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 well-crafted: it states a concrete purpose, supplies natural trigger terms, and explicitly scopes its boundaries against related skills. The only minor gap is that the listed 'actions' are more investigative framing than discrete operations.

Suggestions

Consider listing one or two concrete operations (e.g., 'shortlist suspicious records', 'compare detection algorithms') alongside the investigative framing to strengthen the specificity anchor.

The 'not generic data validation or end-to-end ML ownership' negative phrasing is slightly defensive; the positive boundaries already convey this.

Dimension	Reasoning	Score
Specificity	Names the anomaly-detection domain and several concrete activities ('outliers, rare events, spikes, and suspicious records'), but the actions are framed as investigation/comparison rather than an enumerated set of concrete operations like 'extract', 'fill', 'merge'.	4 / 5
Completeness	Explicitly answers 'what' ('Investigate outliers, rare events, spikes, and suspicious records') and 'when' via a clear 'Use as an explicit anomaly-analysis helper when...' clause with concrete trigger guidance.	5 / 5
Trigger Term Quality	Strong coverage of natural user terms including 'outliers', 'fraud candidates', 'sensor spikes', 'rare failures', and algorithm synonyms (isolation forest, one-class SVM, LOF), matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	A clearly stated niche with explicit boundary routing to sibling skills (exploratory-data-analysis, scikit-learn, scientific-visualization) minimizes conflict risk and signals distinct triggers.	5 / 5
	Total	19 / 20 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
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	14 / 16 Passed

Repository: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/detecting-data-anomalies/SKILL.md
Commit: d5ae560

Reviewed: 1 day 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.