CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dataviz

Use this skill whenever you are about to create ANY chart, graph, plot, dashboard, or data visualization, in ANY output medium (HTML/React artifact, inline SVG, matplotlib/plotly/d3/Recharts, PNG). Read it BEFORE writing the first line of chart code, choosing chart colors, building a stat tile / KPI row, or laying out a dashboard. Teaches a design-system-agnostic method - form heuristic, color formula with a runnable validator, mark specs, interaction rules. Default palette in references/palette.md.

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

85%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.

A well-architected overview: a strictly ordered procedure with an explicit validate-and-fix loop, hard numeric rules, and clean one-level-deep disclosure into a complete, verified bundle. The only weaknesses are mild rule duplication between the procedure and the Non-negotiables section and execution details that require opening the reference files.

Suggestions

Deduplicate rules stated twice — the fixed-order categorical-hue rule (step 2 vs. Non-negotiables) and the legend/direct-label rules (step 6 vs. Non-negotiables) — by stating each once and cross-referencing, tightening conciseness without losing content.

Make the validator command runnable as-is by inlining a one-line example with real hexes from the default palette (e.g. node scripts/validate_palette.js "#4269d0,#efb118,..." --mode light), so the most common execution path works without first opening references/palette.md.

Trim the overlap between the 'Plugging in a design system' paragraph and its parameter table — the prose restates what the table then tabulates.

DimensionReasoningScore

Conciseness

Efficient and dense — rules and pointers with no basic-concept explanations — but with minor trimmable redundancy: the fixed-order categorical rule appears in both step 2 and "Non-negotiables", legend/direct-label rules appear in both step 6 and "Non-negotiables", and the design-system parameter table partially restates the preceding parameterization paragraph. Not 5 because of this repetition; not 3 because there is no padding or explanation of things Claude already knows.

4 / 5

Actionability

Mostly executable: one copy-paste-ready command ("node scripts/validate_palette.js \"<hex,hex,...>\" --mode light") plus concrete numeric specs ("4px rounded data-ends", "2px lines", ">=8px markers", "CVD Delta E >= 8", "normal-vision floor below 15 is a hard FAIL"). Not 5: common-case execution (actual palette hexes, component construction, tooltip markup) is delegated to reference files rather than runnable from the body alone; not 3: what is written is concrete and specific, not high-level hints or pseudocode.

4 / 5

Workflow Clarity

"The procedure - do these in order" gives 7 clearly sequenced steps with an explicit validation checkpoint and feedback loop (step 3: "VALIDATE the palette - run the script... Fix anything that FAILs before continuing", re-run for --mode dark), a final verification step (step 7: "Render it and look at it"), and a closing check against the anti-patterns catalog acting as a checklist. Matches the anchor-5 example's validate -> fix -> retry structure.

5 / 5

Progressive Disclosure

The body is a true overview: each procedure step signals its reference ("-> references/choosing-a-form.md"), and a closing table maps all 7 reference files plus the validator script to the question each answers. Verified against the actual bundle: every referenced path exists and references are flat (they point only to scripts/, no nested chains), so navigation is one level deep and easy.

5 / 5

Total

18

/

20

Passed

Description

100%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.

An exemplar description: it states concrete capabilities, an explicit two-fold 'use when' trigger clause, the method's components, and the palette location, all in dense non-padded prose. Trigger synonyms and format names give it comprehensive natural-language coverage.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities: "create ANY chart, graph, plot, dashboard, or data visualization", "choosing chart colors, building a stat tile / KPI row, or laying out a dashboard", and "form heuristic, color formula with a runnable validator, mark specs, interaction rules" — comprehensive coverage of what the skill does with no evident gaps. Not 4: the action list spans creation, coloring, components, layout, and the method's constituent parts, exceeding 'several specific actions with minor gaps'.

5 / 5

Completeness

Explicitly answers both: what ("Teaches a design-system-agnostic method - form heuristic, color formula with a runnable validator, mark specs, interaction rules") and when ("Use this skill whenever you are about to create ANY chart..." and "Read it BEFORE writing the first line of chart code, choosing chart colors..."). Not 4: the 'when' is not merely present but stated twice with concrete trigger phrases, matching the anchor-5 example structure.

5 / 5

Trigger Term Quality

Comprehensive natural-term coverage with synonyms ("chart, graph, plot, dashboard, or data visualization") plus medium/format equivalents users would say ("HTML/React artifact, inline SVG, matplotlib/plotly/d3/Recharts, PNG", "stat tile / KPI row", "chart colors"). Not 4: no common variation is missing — the synonym set and format/library names function the way file extensions do for the anchor-5 example.

5 / 5

Distinctiveness Conflict Risk

Clear niche (data-visualization design) with distinct, viz-specific triggers (chart, graph, plot, dashboard, stat tile, chart colors); minimal realistic conflict with other skills. Not 4: the trigger vocabulary is tightly scoped to visualization authoring, leaving no meaningful overlap with adjacent skill domains evident from the text.

5 / 5

Total

20

/

20

Passed

Validation

100%

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

JimLiu/baoyu-design

Path

chatgpt-plugins/baoyu-dataviz/skills/dataviz/SKILL.md

Reviewed

Table of Contents

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.