CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

image-file-size

Use when auditing the image pipeline, static assets in `public/`, CMS-managed media, or image CDN transforms. Separate file-size problems from dimension mismatches so you can tell whether the issue is compression, format choice, or responsive delivery.

Quality

74%

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/image-file-size/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%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 delivers concrete, threshold-driven audit and fix guidance and splits implementation detail cleanly to a real one-level reference. It loses points for spending tokens on general web-performance facts that are duplicated in the reference, and for a batch convert workflow that lacks any post-fix verification step.

Suggestions

Cut the Explain section and trim the intro's HTTP Archive / conversion-rate statistics — this is knowledge Claude already has and it duplicates content in references/rule.md.

Add a verification step after Fix: re-check each converted image's file size against the thresholds and confirm Lighthouse no longer reports 'Properly size images' or 'Efficiently encode images'.

Inline one executable example (e.g. a sharp or cwebp one-liner at the documented quality settings) so the Fix steps can be executed without opening the reference.

DimensionReasoningScore

Conciseness

Quick Reference, Check, and Fix are lean and threshold-driven, but the intro ('Images account for over 50% of average page weight... 1-second delay... reduce conversions by up to 20%') and the Explain section ('HTTP Archive data consistently shows images account for 40-60% of total page weight...') restate general web-performance knowledge Claude already has, and duplicate both each other and references/rule.md. Two padded sections is 'some unnecessary explanation', not minor instances.

3 / 5

Actionability

Concrete, executable guidance: specific thresholds (200KB photos, 400KB heroes, 30KB thumbnails), named tools (Squoosh, Sharp, ImageOptim, oxipng, pngquant, SVGO), and quality settings (WebP 80%, AVIF 60%). As an instruction-only skill this is actionable without inline code; the minor gap is that no executable command appears in SKILL.md itself — all code is delegated to references/rule.md.

4 / 5

Workflow Clarity

A clear Check-then-Fix sequence with numbered steps and concrete flag conditions, but no validation checkpoint after Fix (no re-check of file sizes, no Lighthouse re-run to confirm the opportunity cleared). Because Fix batch-rewrites image files (convert, strip metadata), the rubric's cap for batch operations without validation applies, holding this at 3 even though the sequence itself is well-ordered.

3 / 5

Progressive Disclosure

A ~35-line overview with well-organized sections and a clearly signaled one-level-deep reference ('For full implementation details, code examples, and framework-specific guidance, see references/rule.md') that exists in the bundle and appropriately holds the code examples and framework guidance. Content is correctly split and navigation is easy; nothing is deeply nested or buried.

5 / 5

Total

15

/

20

Passed

Description

73%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 strong description with explicit, concrete when-triggers and deliberate boundary-setting against related image skills. Its main weaknesses are a second-person phrase that violates the third-person voice requirement and a thin set of concrete action verbs relative to what the skill actually does (flagging thresholds, reporting, conversion guidance).

Suggestions

Rewrite in third person: replace 'so you can tell whether the issue is compression, format choice, or responsive delivery' with e.g. 'distinguishing whether the issue is compression, format choice, or responsive delivery'.

Enumerate the skill's concrete actions, e.g. 'Flags photos over 200KB, heroes over 400KB, and thumbnails over 30KB; recommends WebP/AVIF conversion' — currently only 'auditing' and 'separating' are named.

Add natural trigger terms users would actually say — 'optimize images', 'images too large', 'page weight' — plus file extensions (.png, .jpg, .webp, .avif).

DimensionReasoningScore

Specificity

The description names the domain ('auditing the image pipeline, static assets in public/, CMS-managed media, or image CDN transforms') and two concrete actions ('Separate file-size problems from dimension mismatches'), matching the anchor for 1-2 concrete actions that are not comprehensive. The rubric's third-person guideline then reduces this by 1 because 'so you can tell whether the issue is compression, format choice, or responsive delivery' uses second-person voice.

2 / 5

Completeness

Explicitly answers 'when' with concrete trigger phrases ('Use when auditing the image pipeline, static assets in public/, CMS-managed media, or image CDN transforms') and 'what' with an explicit functional statement ('Separate file-size problems from dimension mismatches'). Not a 4 because that anchor requires the 'when' to be more explicit or specific, and here the when-clause is fully explicit and leads the description.

5 / 5

Trigger Term Quality

Good keyword coverage with natural phrases users would say ('image pipeline', 'static assets', 'CMS-managed media', 'image CDN transforms', 'compression', 'responsive delivery'). Not a 5: misses common variations like 'optimize images', 'images too large', 'page weight', and file extensions (.png, .jpg, .webp, .avif).

4 / 5

Distinctiveness Conflict Risk

Mostly distinct: the phrase 'Separate file-size problems from dimension mismatches' deliberately carves out a niche against sibling image skills. Not a 5 because broad triggers like 'static assets in public/' and 'image CDN transforms' could still overlap with responsive-image or dimension-related skills.

4 / 5

Total

15

/

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

Path

skills/image-file-size/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.