CtrlK
BlogDocsLog inGet started
Tessl Logo

web-prototype-taste-soft

Apple-tier soft web prototype. Silver/cream canvas, double-bezel cards, button-in-button CTAs, generous squircle radii, spring motion, ambient mesh. Distilled from Leonxlnx/taste-skill `soft-skill` + sections 4–8 of `taste-skill`.

60

Quality

68%

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

Fix and improve this skill with Tessl

tessl review fix ./design-templates/web-prototype-taste-soft/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%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 body is an unusually dense, highly actionable rulebook — exact values, Tailwind classes, and a validation checklist — with essentially no padding. Its main flaws are a dangling `example.html` reference and the absence of an explicit build sequence or fix-loop behind the pre-flight checklist.

Suggestions

Ship the referenced `example.html` in the skill directory (or drop the reference) — as written, "See `example.html` in this directory" points at a file that does not exist in the bundle.

Add one line of build order and a fix loop, e.g. "Assemble in page order (nav → hero → bento → feature → marquee → closing → footer), then run the pre-flight checklist; fix any failing item before delivery."

Fold the "Source" section's distillation meta-commentary into a single provenance line (or into frontmatter) — the body already carries the full rule set, so the section mostly repeats what the description states.

DimensionReasoningScore

Conciseness

The rule format is dense and assumes competence — "**Canvas:** silver-grey or warm cream (`#F2F2F0` or `#FDFBF7`). Not pure white." — with no explanation of concepts Claude already knows. Not a 5: the "Source" section ("Distilled from Leonxlnx/taste-skill — `skills/soft-skill/SKILL.md` (\"Vanguard UI Architect\") with the haptic micro-aesthetics...") duplicates provenance already in the frontmatter and is trimmable meta-commentary.

4 / 5

Actionability

Fully copy-paste-ready values throughout: `clamp(48px, 7vw, 96px)`, `letter-spacing: -0.035em`, `border-radius: 28px–40px`, `cubic-bezier(0.32, 0.72, 0, 1)`, `rounded-full px-6 py-3`, `0 20px 40px -15px rgba(0,0,0,0.05)`, `translateX(-50%)` over `30s linear`, plus exact Tailwind classes and banned-pattern replacements. The common cases (color, type, radii, shadows, motion, nav, hero, grid) are each covered with specifics.

5 / 5

Workflow Clarity

The "Required components" list reads as a page-assembly sequence (navbar → asymmetric hero → bento grid → feature row → marquee → closing band → footer), and the "Pre-flight" checklist is an explicit validation checkpoint. Not 5: there is no stated build order or feedback loop for what to do when a pre-flight item fails; not 3: a concrete validation checklist is present, and nothing here is a destructive/batch operation that would trigger the cap.

4 / 5

Progressive Disclosure

Sections are well-organized and the content is appropriately inline for a dense rule set, but the body's only bundle reference — "See `example.html` in this directory for a fully-realized template" — points to a file that does not exist in the bundle (no `references/`, `scripts/`, `assets/`, or `example.html` alongside SKILL.md). Per the rubric's guidance to score against the actual bundle structure, this is a broken navigation promise rather than a clearly signaled one-level-deep reference. Not 4: a dangling reference sends Claude down a dead end, which is more than a minor organization gap.

3 / 5

Total

16

/

20

Passed

Description

58%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.

The description is distinctive and concrete about the visual vocabulary, but reads as a feature inventory rather than a capability-plus-trigger statement. It lacks any "Use when..." guidance, which caps completeness and leaves trigger-term coverage thin.

Suggestions

Add an explicit trigger clause, e.g. "Use when a brief asks for Apple-like, Linear-tier, premium consumer, or calm SaaS polish on a web page or landing page."

Lead with a third-person action statement ("Builds soft premium web prototypes...") so the description answers "what it does", not just "what it looks like".

Move the "Distilled from Leonxlnx/taste-skill..." provenance detail out of the description into the body — it consumes trigger-word budget without helping a user decide to invoke the skill.

DimensionReasoningScore

Specificity

Lists several concrete, distinctive elements — "double-bezel cards", "button-in-button CTAs", "generous squircle radii", "spring motion, ambient mesh", plus concrete palette values ("Silver/cream canvas") — going beyond the 1–2 items of anchor 3. Not a 5: it is a style-feature inventory rather than a comprehensive statement of the actions the skill performs (no "builds/generates/apply" verbs), so coverage of capabilities has gaps.

4 / 5

Completeness

The "what" is reasonably clear (a soft-premium web prototype aesthetic with an enumerated feature set), but there is no "when" — no "Use when..." clause or equivalent trigger guidance, which caps completeness at 3 per the rubric guideline. Not 4: the "when" is entirely absent rather than merely improvable.

3 / 5

Trigger Term Quality

Some relevant keywords exist ("Apple-tier", "soft web prototype", "premium" via implication), but the natural phrases a user would actually say are missing or thin — no "Apple-like", "Linear-style", "landing page", "premium SaaS", or "Use when..." trigger phrasing. Not 4: the keyword set lacks common variations and synonyms; not 2: the terms present are domain-relevant, not generic filler.

3 / 5

Distinctiveness Conflict Risk

Niche vocabulary ("double-bezel", "button-in-button", "squircle radii", "ambient mesh") gives it a clear identity with minimal overlap against generic skills. Not 5: it self-identifies as "Distilled from Leonxlnx/taste-skill `soft-skill`", so sibling aesthetic skills derived from the same parent (and broad "premium/Apple-like" phrasing) carry minor collision risk.

4 / 5

Total

14

/

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
nexu-io/open-design
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.