explore-data

Profile and explore a dataset to understand its shape, quality, and patterns. Use when encountering a new table or file, checking null rates and column distributions, spotting data quality issues like duplicates or suspicious values, or deciding which dimensions and metrics to analyze.

Quality

77%

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 ./data/skills/explore-data/SKILL.md

Quality

Content

63%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 delivers actionable profiling guidance with specific thresholds and executable SQL inside a clear 7-step workflow, but it is a monolithic 320-line document. Secondary reference material is inlined instead of split into references/ files, textbook statistical concepts pad the token budget, and the one external reference is a broken path.

Suggestions

Move the Quality Assessment Framework, Pattern Discovery Techniques, and Schema Understanding/Documentation sections into references/ files (e.g. references/quality.md, references/schema-docs.md), keeping SKILL.md as a lean workflow overview with clearly signaled one-level-deep links.

Trim or delete concept definitions Claude already knows (distribution shapes like normal/skewed/bimodal, trend/seasonality glossary, "correlation does not imply causation") and de-duplicate the placeholder-value list that appears in both step 4 and the Accuracy Indicators section.

Fix or remove the broken ../../CONNECTORS.md link, or replace it with a short inline note about checking connected tools.

Dimension	Reasoning	Score
Conciseness	The 7-step workflow is directive and efficient, but the back half pads with textbook statistics Claude already knows ("Skewed right: Long tail of high values", "Bimodal: Two peaks (suggests two distinct populations)", "Correlation does not imply causation") and repeats material (the placeholder-value list "999999", "N/A", "TBD", "test" appears in both step 4 and Accuracy Indicators). Mostly efficient overall, so above the verbose anchor 2 but not the lean anchor 4.	3 / 5
Actionability	Concrete, executable guidance dominates: specific thresholds (>5% warn / >20% alert nulls, \|r\| > 0.7 correlation flag, cardinality 3-50), exact percentile lists (p1, p5, p25, p75, p95, p99), and runnable PostgreSQL information_schema queries. Minor gaps: warehouse-specific SQL is PostgreSQL-only with no Snowflake/BigQuery variants, so short of anchor 5.	4 / 5
Workflow Clarity	A clearly sequenced 7-step workflow with numbered sub-steps, explicit fallbacks when neither warehouse nor file is available, and a concrete output template. Read-only profiling triggers no destructive-operation cap, but there are no explicit verification checkpoints or feedback loops, so it sits at anchor 4 rather than 5.	4 / 5
Progressive Disclosure	Well-headered single file, but at ~320 lines secondary material that belongs in reference files (schema documentation template, schema exploration SQL, pattern discovery techniques, quality assessment framework) is fully inlined rather than split out, and the only external pointer, ../../CONNECTORS.md, does not exist in the bundle. Structure is present but content that should be separate is inline, matching anchor 3 rather than 4.	3 / 5
	Total	14 / 20 Passed

Description

83%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.

A strong description: concrete third-person actions, an explicit and multi-phrase Use-when clause, and a clear profiling niche. It falls just short of top marks on specificity and trigger coverage because several body capabilities and common file-extension synonyms are absent.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("checking null rates and column distributions", "spotting data quality issues like duplicates or suspicious values", "deciding which dimensions and metrics to analyze") in third-person voice, but coverage of the skill's full capability set (schema documentation, correlation analysis, follow-up recommendations) is incomplete.	4 / 5
Completeness	Explicitly answers both what ("Profile and explore a dataset to understand its shape, quality, and patterns") and when ("Use when encountering a new table or file, checking null rates and column distributions..."), with concrete trigger phrases throughout the Use-when clause.	5 / 5
Trigger Term Quality	Natural phrases like "encountering a new table or file", "null rates", "duplicates", and "suspicious values" mirror what users would say, but common variations and file extensions (.csv, .xlsx, "data profiling") are missing.	4 / 5
Distinctiveness Conflict Risk	Data profiling is a distinct niche with specific triggers (null rates, column distributions, duplicates) that are unlikely to fire for unrelated skills; minor overlap remains with general data-analysis skills because "dataset"/"table" are broad nouns.	4 / 5
	Total	17 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: anthropics/knowledge-work-plugins
Path: data/skills/explore-data/SKILL.md
Commit: 8444efc

Reviewed: 6 days 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.