CtrlK
BlogDocsLog inGet started
Tessl Logo

unused-css

Use when reviewing stylesheets, component styles, and responsive behavior related to Remove unused CSS rules. Check the rendered layout across breakpoints and interaction states before proposing a fix.

57

Quality

66%

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 ./skills/unused-css/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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 well-structured with exemplary progressive disclosure — a short overview deferring cleanly to an existing references/rule.md — but the substance is thin: no executable commands or configuration in the SKILL.md itself, no validation step for a batch purge operation that can break rendering, and some redundancy (duplicate tool mentions, an Explain section restating common knowledge).

Suggestions

Include one concrete, copy-paste-ready example in the body, e.g. an npx purgecss command or a PurgeCSS/Vite config snippet with a safelist, so the skill is actionable without opening the reference.

Add an explicit validation step to the Check/Fix flow (e.g. re-check the rendered layout across breakpoints and interaction states after purging, before committing) to satisfy the feedback-loop requirement for batch CSS removal.

Tighten redundancy: merge the "Check" section with "Quick Reference" (both name PurgeCSS/UnCSS) and cut or compress the "Explain" section, which describes concepts Claude already knows.

DimensionReasoningScore

Conciseness

The body is short and mostly lean, but contains redundant and low-value content that could be tightened: PurgeCSS/UnCSS are named in both "Quick Reference" and "Check" within a few lines, and the "## Explain" section ("Explain how unused CSS bloats bundle sizes, slows page loading...") tells Claude what to explain about concepts it already knows. This places it at the 'mostly efficient but some unnecessary explanation' anchor rather than the level-4 'only minor instances could be trimmed', and well above level 2 since there is no padded tutorial-style prose.

3 / 5

Actionability

"Use PurgeCSS or UnCSS to remove unused selectors" and "Integrate into build pipeline (Webpack, Vite, PostCSS)" name real tools and ecosystems, which is concrete direction, but the body contains no executable command, config snippet, or purge configuration — the actionable specifics are entirely deferred to references/rule.md. This matches the level-3 anchor (some concrete guidance but incomplete, missing key details) rather than level 4, which requires actual code or commands with only minor gaps.

3 / 5

Workflow Clarity

There is a rough implicit sequence (Quick Reference → Check → Fix) with labeled sections, but no explicit step ordering and no validation checkpoints. Because purging CSS is a batch operation that can break the rendered UI, the rubric caps workflow clarity at 3; the only safety nod is "Safelist dynamic classes and framework-specific patterns", with no verify-rendered-output step despite the frontmatter promising to "Check the rendered layout across breakpoints and interaction states before proposing a fix".

3 / 5

Progressive Disclosure

The body is a ~45-line overview with well-organized sections and a single clearly-signaled, one-level-deep reference — "For full implementation details, code examples, and framework-specific guidance, see references/rule.md" — and that file exists in the bundle (references/ contains exactly rule.md, with no scripts/ or assets/). The split (overview in SKILL.md, full details in the reference) matches the level-5 anchor for a short, well-organized skill.

5 / 5

Total

14

/

20

Passed

Description

75%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.

A solid description with an explicit 'Use when' trigger clause and several concrete review actions, scoring uniformly at the 'good but not exemplary' level. Its main weakness is the ungrammatical construction "related to Remove unused CSS rules", which obscures the skill's core purpose and leaves natural synonyms (dead CSS, purge, bundle size) out of the trigger terms.

Suggestions

Rewrite the trigger phrase into a declarative what + when structure, e.g. "Identify and remove unused CSS rules with PurgeCSS/UnCSS to reduce bundle size. Use when reviewing stylesheets, component styles, or responsive behavior for dead or unused CSS."

Add natural trigger synonyms such as "dead CSS", "CSS bloat", "purge CSS", and ".css files" to broaden the terms users would actually say.

State the concrete outcome (typical 50-90% CSS reduction, faster page loads) explicitly so the 'what' does not depend on the awkward "related to Remove unused CSS rules" clause.

DimensionReasoningScore

Specificity

The description lists several concrete actions — "reviewing stylesheets, component styles, and responsive behavior", "Check the rendered layout across breakpoints and interaction states", "before proposing a fix" — which goes beyond the 1-2 actions of the level-3 anchor. It is not level 5 because the core capability (identifying/removing unused CSS with specific tooling) is only alluded to via the awkward phrase "related to Remove unused CSS rules", leaving a minor coverage gap.

4 / 5

Completeness

Both parts are present: an explicit trigger clause ("Use when reviewing stylesheets, component styles, and responsive behavior...") and a what-portion ("Check the rendered layout across breakpoints and interaction states before proposing a fix"). It falls short of level 5 because the 'what' is muddled — the skill's actual purpose (removing unused CSS to cut bundle size) is buried in the ungrammatical "related to Remove unused CSS rules" rather than stated as a clear action, and it sits above level 3 since the 'when' is explicit, not weakly implied.

4 / 5

Trigger Term Quality

Good keyword coverage with natural terms users would say: "stylesheets", "component styles", "responsive behavior", "breakpoints", "interaction states", "unused CSS rules". Not level 5 because common synonyms and variations are missing ("dead CSS", "CSS bloat", "purge CSS", ".css", "bundle size"), and not level 3 because coverage clearly exceeds a single generic keyword like "works with CSS files".

4 / 5

Distinctiveness Conflict Risk

The unused-CSS niche is identifiable ("Remove unused CSS rules", "stylesheets"), giving it a mostly distinct trigger profile with minor overlap risk against general CSS/layout-review skills. Not level 5 because the leading triggers ("reviewing stylesheets, component styles, and responsive behavior") are broad enough that a general stylesheet-review request could plausibly invoke it; not level 3 because the unused-CSS marker clearly separates it from generic CSS skills.

4 / 5

Total

16

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository
thedaviddias/Front-End-Checklist
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.