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

70%

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

50%

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

The content is admirably concise and well-structured as a positioning overview, but it lacks actionable, executable guidance — it describes the skill's scope instead of instructing how to perform anomaly detection. Bundle files exist but are neither referenced from the body nor accurately inventoried in their own README.

Suggestions

Add a concrete worked example or a minimal executable snippet (e.g., a scikit-learn IsolationForest call with a contamination threshold) so the body instructs rather than only describes.

Wire the existing bundle into the body with one-level-deep links (e.g., "See scripts/algorithm_selector.py" and "See references/implementation.md") so progressive disclosure actually routes to the detailed material.

Fix references/README.md so it lists only real files (errors.md, examples.md, implementation.md) and remove the phantom anomaly_detection_algorithms.md / data_preprocessing_guide.md / performance_metrics.md / case_studies.md entries.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — short sections like "When to Use", "Not For / Boundaries", and "Typical Outputs" carry no concept explanations or padding, so every token earns its place.	3 / 3
Actionability	Guidance is descriptive rather than instructive — "Reviewing outlier transactions", "Comparing isolation forest, one-class SVM, LOF, or threshold-based anomaly workflows", "Turning suspicious records into a shortlist" — with no executable code, commands, thresholds, or concrete steps, matching the score-1 "describes rather than instructs" anchor.	1 / 3
Workflow Clarity	Sections give a loose Positioning → When to Use → Not For → Outputs structure, but there is no numbered multi-step sequence and no validation/verification checkpoints for the batch anomaly operations the bundled scripts imply; the missing feedback loop caps this at 2.	2 / 3
Progressive Disclosure	The body is a well-organized short overview, but it never signals the existing references/ or scripts/ bundle files, and references/README.md lists files (anomaly_detection_algorithms.md, data_preprocessing_guide.md, etc.) that do not exist in the directory, leaving the bundle partly phantom and unwired.	2 / 3
	Total	8 / 12 Passed

Description

90%

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 third-person, concise, and clearly states both what the skill does and when to use it, with strong natural trigger terms and explicit boundary language. The only soft spot is specificity, where the action list leans on target record types rather than distinct concrete operations.

Dimension	Reasoning	Score
Specificity	Phrases like "Investigate outliers, rare events, spikes, and suspicious records" name the domain and several record types, but read as a list of targets rather than multiple distinct concrete operations, so they stop short of the score-3 anchor.	2 / 3
Completeness	It states both what ("Investigate outliers, rare events, spikes, and suspicious records") and an explicit when ("Use as an explicit anomaly-analysis helper when you want concrete anomaly-detection workflow guidance"), satisfying the score-3 anchor.	3 / 3
Trigger Term Quality	Natural user-facing terms are well covered — "outliers, rare events, spikes, and suspicious records" plus "anomaly-detection workflow guidance" — matching the score-3 anchor of good coverage of words users would actually say.	3 / 3
Distinctiveness Conflict Risk	The niche is clear and fenced off — "not generic data validation or end-to-end ML ownership" and "anomaly-detection ownership normally belongs to scikit-learn" — making it unlikely to trigger for the wrong skill.	3 / 3
	Total	11 / 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
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: f627ab5

Reviewed: about 8 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.