data-quality

Quality dimensions quick reference and remediation decision tree for tabular data assessment

Quality

65%

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 ./.claude/skills/skills/data-quality/SKILL.md

Quality

Content

80%

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

The body is an exceptionally lean and actionable quick-reference, with concrete qsv commands and thresholds throughout; its main gaps are the absence of explicit post-fix validation checkpoints in the fix workflow and a lack of progressive-disclosure structure for the large advanced-stats reference table.

Suggestions

Add an explicit verification step to the Fix Ordering workflow (e.g. '7. Re-run `stats --cardinality` / `validate` to confirm the issue is resolved before proceeding') to close the validate->fix->retry loop for destructive fixes.

Consider moving the Advanced Stats metrics table into a references/ file (e.g. ADVANCED_STATS.md) and linking to it from the main body, giving the skill a clearer one-level-deep progressive-disclosure structure.

Dimension	Reasoning	Score
Conciseness	The body is dense reference material — dimension, remediation, fix-ordering, and stats-cache tables — with no padding or explanation of concepts Claude already knows; every row carries a concrete signal and tokens earn their place.	3 / 3
Actionability	Concrete, executable qsv commands appear throughout (e.g. `stats --cardinality`, `validate schema.json`, `sqlp with TRIM(col)`, `moarstats --advanced`), with specific thresholds and copy-paste-ready fix recipes.	3 / 3
Workflow Clarity	The 'Fix Ordering' section gives a clear numbered sequence and the remediation table's 'When to Skip' column provides implicit gating, but there is no explicit validate-then-retry checkpoint after applying destructive fixes (dedup, fixlengths, sqlp transforms), so the top anchor is not reached.	2 / 3
Progressive Disclosure	The file is well-organized into clear sections, but it is a single monolithic reference with no bundle files and no one-level-deep references; the substantial 'Advanced Stats' table is inline content that could plausibly live in a separate reference, so it sits at the middle anchor.	2 / 3
	Total	10 / 12 Passed

Description

50%

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 conveys a clear domain and two deliverables but omits any explicit 'when to use' trigger and relies on somewhat technical phrasing, leaving it short of the top anchor on every dimension.

Suggestions

Add an explicit trigger clause, e.g. 'Use when assessing or remediating quality issues in tabular/CSV data — missing values, duplicates, invalid formats, or orphaned references.'

Include natural user-facing terms ('data quality', 'bad data', 'missing values', 'duplicates') alongside the technical vocabulary to improve trigger_term_quality.

Sharpen distinctiveness by naming the qsv tooling context so it does not blur into the /data-profile workflow.

Dimension	Reasoning	Score
Specificity	Names the domain ('tabular data assessment') and two concrete artifacts ('quality dimensions quick reference' and 'remediation decision tree'), but does not enumerate multiple discrete actions like the score-3 example; it describes deliverables rather than listing specific actions.	2 / 3
Completeness	It states what the skill provides (quick reference + decision tree) but has no 'Use when...' clause or equivalent explicit trigger guidance; per the guidelines a missing explicit trigger caps completeness at 2.	2 / 3
Trigger Term Quality	Relevant keywords exist ('quality dimensions', 'remediation', 'tabular data assessment') but they lean technical; common natural phrasings a user would say ('data quality', 'missing values', 'duplicates', 'fix bad data') are absent, so coverage is partial.	2 / 3
Distinctiveness Conflict Risk	The 'tabular data assessment' niche is reasonably specific, but the phrasing could overlap with the related /data-profile workflow and other data skills; it is not uniquely anchored by distinct triggers.	2 / 3
	Total	8 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: dathere/qsv
Path: .claude/skills/skills/data-quality/SKILL.md
Commit: 17c10db

Reviewed: about 5 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.