profiling-tables

Deep-dive data profiling for a specific table. Use when the user asks to profile a table, wants statistics about a dataset, asks about data quality, or needs to understand a table's structure and content. Requires a table name.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

72%

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

A well-structured, code-heavy profiling recipe that is lean and clearly sequenced, but its SQL templates are not fully executable as written and it lacks validation checkpoints for batch database work.

Suggestions

Mark column placeholders consistently (e.g. '<column_name>') and show how to iterate the column-stat queries across all columns, so the examples are copy-paste ready.

Add explicit validation checkpoints — e.g. confirm the table resolves and spot-check that aggregate queries return sensible rows before running the full column-stat pass — with a fix-and-retry loop.

Clarify the run_sql execution context for Step 2 onward so Claude knows where each query is meant to be submitted.

Dimension	Reasoning	Score
Conciseness	The body is dominated by lean SQL templates and guiding questions, assuming Claude's competence without explaining what SQL, NULLs, or percentiles are; nearly every token advances the task.	3 / 3
Actionability	Real SQL is provided throughout, but queries use placeholders like '<table>' alongside the literal token 'column_name' that is not clearly marked as a placeholder, and the 'for each column' iteration is left implicit, so examples are not copy-paste executable.	2 / 3
Workflow Clarity	A clear seven-step sequence is present, but for a database/batch operation there are no explicit validation checkpoints or error-recovery feedback loops beyond a single locate-the-table hint, which caps this dimension at 2.	2 / 3
Progressive Disclosure	No bundle files exist and none are needed; the content is a single cohesive workflow organized into clearly labeled numbered steps and subsections, with no nested references.	3 / 3
	Total	10 / 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.

A concise, well-constructed description that clearly conveys both capability and trigger conditions in third person. It hits every dimension the rubric rewards without padding.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — profiling a table, gathering statistics, assessing data quality, and understanding a table's structure and content — rather than vague language.	3 / 3
Completeness	Explicitly states what it does ('Deep-dive data profiling for a specific table') and when to use it via a clear 'Use when...' clause with multiple triggers, plus a precondition ('Requires a table name').	3 / 3
Trigger Term Quality	Natural user-facing phrases like 'profile a table', 'statistics about a dataset', 'data quality', and 'understand a table's structure and content' cover the ways a user would actually request this.	3 / 3
Distinctiveness Conflict Risk	The table-profiling niche and the 'Requires a table name' constraint give it distinct triggers unlikely to conflict with unrelated skills.	3 / 3
	Total	12 / 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: astronomer/agents
Commit: 8827e93

Reviewed: 8 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.