CtrlK
BlogDocsLog inGet started
Tessl Logo

liquid-metal-border

Add and tune animated liquid-metal WebGL borders with the React `metal-fx` package. Use when buttons, icon controls, chips, tabs, cards, or selected surfaces need a metallic active, selected, hover, focus, or premium border; when implementing the MetalFx component from metal.jakubantalik.com; or when troubleshooting its presets, themes, strength, glow, reflections, sizing, radius, animation, accessibility, SSR, or performance.

75

Quality

92%

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 strong, dense body: highly actionable code, a complete prop surface, explicit verification and failure-recovery loops, and almost no filler — its one padding spot is generic React boilerplate inside the active-state example. Its main structural weakness is that everything lives in SKILL.md; the prop reference, tuning defaults, and failure diagnosis would earn their keep as one-level-deep reference files loaded on demand.

Suggestions

Move the 'Complete Customization Surface' prop table, 'Tuning Defaults', and 'Failure Diagnosis' into a references/ file (e.g. references/props.md and references/troubleshooting.md) with clearly signaled one-level-deep links from SKILL.md, keeping the core contract, baseline patterns, and verification checklist inline.

Trim the active-state example by replacing the full `useReducedMotion` hook and hover/focus state wiring with a brief mention — show only the `MetalFx` prop wiring (`strength`, `paused`, `disableGlow`, `ringCssPx`) that is specific to this package.

The 'Handoff' section overlaps the 'Verification' checklist's reporting intent; merge the two so verification results feed a single handoff step instead of two parallel closing lists.

DimensionReasoningScore

Conciseness

Nearly every section carries package-specific knowledge (prop semantics, dark-mode-only reflections, shared WebGL context, SSR behavior) that Claude cannot know, and the prose is terse. It is not 5 because the ~50-line active-state example embeds generic React boilerplate Claude already knows — the full `useReducedMotion` hook and hover/focus state wiring could be reduced to a one-line mention ("use a prefers-reduced-motion hook") with only the `MetalFx` prop wiring shown. Not 3 because the padding is localized to one example, not spread across sections.

4 / 5

Actionability

Install commands for four package managers, copy-paste-ready TSX/CSS for the baseline button, active-state toggle, card, sizing, and reflection patterns, a complete prop table with types and defaults, numeric tuning defaults ("idle strength `0.1–0.2`, selected strength `0.75–0.9`"), and symptom→fix diagnosis entries. Fully executable guidance covering the common cases; no pseudocode.

5 / 5

Workflow Clarity

Clear sequence: "Inspect Before Editing" (pre-flight checks including version check), install, baseline pattern, then an 11-item "Verification" checklist with explicit checkpoints (reduced motion, breakpoints, light/dark, offscreen resume), followed by "Failure Diagnosis" entries that pair each symptom with its fix — a validate/diagnose/retry feedback loop. This is not a destructive or batch skill, so no validation cap applies; not 4 because both the verification checklist and the error-recovery mapping are explicit rather than implied.

5 / 5

Progressive Disclosure

The body is well-sectioned but entirely monolithic at ~310 lines with no bundle files: the ~20-line "Complete Customization Surface" prop table, "Tuning Defaults", and "Failure Diagnosis" are exactly the bulk reference material the anchor 3 example says "could be in a separate file" (e.g. references/props.md, references/troubleshooting.md), with the core contract and baseline patterns staying in SKILL.md. Not 4 because there is no offload at all — every level of detail is inlined and always loaded; not 2 because section headers keep it navigable.

3 / 5

Total

17

/

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 exemplary description: third-person voice, concrete actions, an explicit multi-clause 'Use when...' trigger list covering both surface vocabulary and exact technical identifiers, and a tightly scoped niche. Both what and when are answered without padding or buzzwords.

DimensionReasoningScore

Specificity

"Add and tune animated liquid-metal WebGL borders with the React `metal-fx` package" states concrete actions (add, tune, troubleshoot), and the clause "troubleshooting its presets, themes, strength, glow, reflections, sizing, radius, animation, accessibility, SSR, or performance" comprehensively enumerates the tuning surface. Not 4 because coverage of what the skill manipulates is exhaustive for this niche, not just 'several actions with minor gaps'.

5 / 5

Completeness

The 'what' is explicit ("Add and tune animated liquid-metal WebGL borders with the React `metal-fx` package") and the 'when' is explicit with three concrete trigger clauses introduced by "Use when...". This matches the anchor 5 example structure (action statement followed by 'Use when...' with concrete triggers); nothing is only implied.

5 / 5

Trigger Term Quality

Natural user phrasing is thoroughly covered: "buttons, icon controls, chips, tabs, cards", "metallic active, selected, hover, focus, or premium border", plus the package name "metal-fx", the component name "MetalFx", and the site "metal.jakubantalik.com" as direct lookup triggers. Not 4 because both colloquial UI vocabulary and exact technical identifiers are present; no common synonym family is missing.

5 / 5

Distinctiveness Conflict Risk

The description is pinned to a unique package ("metal-fx", "MetalFx component from metal.jakubantalik.com") and a narrow visual effect ("animated liquid-metal WebGL borders"), so it occupies a clear niche with minimal overlap with generic CSS/border or animation skills. Not 4 because no closely related skill would plausibly intercept these triggers.

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
MengTo/Skills
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.