visual-data-dictionary

Build a Visual Data Dictionary — an interactive qsv viz smart dashboard driven by an LLM-inferred JSON Schema data dictionary, with the dictionary browsable beside the charts. Use when the user asks for a visual data dictionary, a documented dashboard, a dictionary-driven dashboard, or wants to explore and document a CSV at the same time. Optionally bins rows into GeoJSON regions.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

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

The body is highly actionable with a clear, validated multi-stage workflow, but it leans long with some over-explained passages and a monolithic structure. The most concrete defect is a broken reference: edit_dictionary.py is cited as a bundle file beside SKILL.md but does not exist in the skill directory.

Suggestions

Provide the missing edit_dictionary.py (Stage 2.5 and Guardrails reference it as "beside this SKILL.md" / "$SKILL_DIR/edit_dictionary.py", but no such file exists in the skill bundle) or remove the references.

Move the ~80-line inline GeoJSON feature-id-key inspector into a scripts/ file and reference it, reducing the SKILL.md body length and improving progressive disclosure.

Tighten the repo-local packaging rationale and de-duplicate the out-of-band TUI warning that appears in both Stage 2.5 and the Guardrails section.

Dimension	Reasoning	Score
Conciseness	Per-line it is tight and free of basic-concept padding, but the ~480-line body and several over-explained passages — the repo-local packaging rationale, the inline ~80-line GeoJSON inspector, and the out-of-band TUI warning repeated in both Stage 2.5 and Guardrails — could be tightened, matching the "mostly efficient but could be tightened" anchor.	2 / 3
Actionability	Provides fully executable, copy-paste-ready bash and python — the denull/describegpt/viz-smart invocations with real flags, inline verification scripts, and concrete remediation branches — matching the fully-executable anchor rather than pseudocode.	3 / 3
Workflow Clarity	A clearly ordered five-stage pipeline ("in this order and no other") with explicit validation checkpoints (Stage 0 preconditions, the dictionary coverage check, Stage 5's test -s and python verification) and feedback loops for the destructive denull step, satisfying the highest anchor including for batch/destructive operations.	3 / 3
Progressive Disclosure	It has clear section/stage structure, but no references/, scripts/, or assets/ bundle directories exist and the body's referenced edit_dictionary.py ("beside this SKILL.md") is absent — a broken reference — while the large GeoJSON inspector and TUI helper are inlined rather than split into referenced files, matching the "content that should be separate is inline" anchor and falling short of the well-split overview.	2 / 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 strong, specific description that states concrete capabilities, includes a natural-language "Use when" trigger clause, and occupies a clear niche unlikely to conflict with other skills. Third-person voice is maintained throughout. No changes needed.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions in third person — "Build a Visual Data Dictionary," "qsv viz smart dashboard driven by an LLM-inferred JSON Schema data dictionary," "dictionary browsable beside the charts," "bins rows into GeoJSON regions" — matching the anchor for multiple specific concrete actions rather than naming a domain with only some actions.	3 / 3
Completeness	It explicitly answers both what (build the dashboard with the dictionary beside charts) and when ("Use when the user asks for a visual data dictionary..."), satisfying the highest anchor with an explicit trigger clause rather than leaving the when implied.	3 / 3
Trigger Term Quality	The "Use when" clause surfaces natural phrasings a user would say — "visual data dictionary," "documented dashboard," "dictionary-driven dashboard," "explore and document a CSV" — giving good coverage of common variations, not a single keyword.	3 / 3
Distinctiveness Conflict Risk	The narrow niche — a qsv-driven, dictionary-embedded viz dashboard with optional GeoJSON binning — has distinct triggers unlikely to fire for generic dashboard or charting skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: dathere/qsv
Path: .claude/skills/visual-data-dictionary/SKILL.md
Commit: e5a3bdc

Reviewed: about 6 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.