CtrlK
BlogDocsLog inGet started
Tessl Logo

webgl-experience

A full-screen, real-time WebGL/WebGL2 experience — animated shaders, 3D scenes, generative visuals, particle fields — rendered live on the GPU with a typographic overlay. Produced as a single self-contained `index.html`. Use when the brief asks for a "WebGL", "shader", "3D", "generative", "GPU", "interactive canvas", "hero animation", or "real-time visual" experience. OpenDesign serves this in powered-preview mode so the GPU stack actually runs.

68

Quality

83%

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

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.

A tight, well-structured body with a clear technique-selection menu, concrete GPU hygiene rules, and a validation checklist — the writing itself is near-exemplary for an instruction-only skill. Its one material defect is bundle-level: the workflow's Step 0 and resource map both point at example.html, which does not exist in the skill directory, breaking the primary detailed reference the content depends on.

Suggestions

Add the missing example.html (the referenced 'working WebGL2 aurora-field reference') to the bundle, or remove the resource-map and Step 0 references to it and inline a minimal shader boilerplate snippet instead.

Close the feedback loop in Step 4: after the self-review checklist, add an explicit 'if any box fails, fix and re-verify before delivering' instruction so the checklist drives re-validation rather than just final inspection.

Inline a short canonical skeleton (canvas + fullscreen-triangle draw loop + DPR-aware resize) so the core pattern is executable from SKILL.md even if a reference file is unavailable.

DimensionReasoningScore

Conciseness

The ~45-line body is lean: no 'what is WebGL' padding, no library tutorials, and the one conceptual section ('Why this is a powered artifact') conveys platform-specific OpenDesign sandbox facts Claude cannot know, so every token earns its place. It is not 4 because there is no section that reads as trimmable over-explanation.

5 / 5

Actionability

Concrete, executable specifics are present — 'alpha:false, DPR clamped to 2', 'getShaderInfoLog', the four named techniques, 'Headline ≤ 14ch', and a described canvas/draw-loop/resize structure. It is not 5 because the body contains no code snippet and its working reference is external; it is not 3 because the guidance given is specific and directly executable for an instruction-only skill, with only a minor gap left to the (missing) reference file.

4 / 5

Workflow Clarity

Steps 0–4 are clearly sequenced and end in an explicit P0 self-review checklist (60fps, clean getError(), resize crispness, contrast, graceful degradation), plus build-time error logging via getShaderInfoLog. It is not 5 because there is no explicit 'if a checklist item fails, fix and re-verify' recovery loop; it is not 3 because validation checkpoints are explicit, not merely implied.

4 / 5

Progressive Disclosure

Sections are well organized and the reference is loudly signaled ('READ FIRST', a resource map), but the bundle does not actually contain example.html — the single referenced file is missing, so the promised one-level-deep detail cannot be loaded. Per the guideline to score against the actual bundle structure, this broken reference leaves the disclosure structure not better than 'some structure' despite the clean sectioning.

3 / 5

Total

16

/

20

Passed

Description

88%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: it states a clear, multi-faceted 'what', an explicit 'Use when' clause with quoted trigger phrases, and stays in third-person declarative voice with no fluff. The only weaknesses are a few missing natural synonyms (GLSL, fragment shader, shader background) and slightly broad triggers ('hero animation', 'interactive canvas') that create minor overlap risk with general animation skills.

DimensionReasoningScore

Specificity

Multiple concrete capabilities are enumerated — 'animated shaders, 3D scenes, generative visuals, particle fields — rendered live on the GPU with a typographic overlay' — plus the concrete output form 'a single self-contained index.html'. This matches the comprehensive multi-action anchor; it is not score 4 because coverage of the domain's visual families is essentially complete rather than having minor gaps.

5 / 5

Completeness

It explicitly answers both questions: the 'what' is the full-screen real-time WebGL/WebGL2 experience with typographic overlay in one index.html, and the 'when' is an explicit 'Use when the brief asks for a... experience' clause with concrete trigger phrases — a direct match to the 5 anchor's exemplar structure. It is not below 5 because neither half is vague or implied.

5 / 5

Trigger Term Quality

The 'Use when' clause quotes eight natural trigger phrases ('WebGL', 'shader', '3D', 'generative', 'GPU', 'interactive canvas', 'hero animation', 'real-time visual') plus 'WebGL2' earlier — good coverage, but common synonyms users actually say such as 'GLSL', 'fragment shader', 'shader background/art', or 'three.js' are absent, so it sits between the 4 and 5 anchors rather than clearly at 5.

4 / 5

Distinctiveness Conflict Risk

The core niche (GPU/WebGL real-time visuals, 'raymarch', 'particle field') is distinct with minimal conflict, matching the 'mostly distinct' anchor. It is not 5 because broader quoted triggers like 'hero animation' and 'interactive canvas' could also plausibly match non-WebGL animation or 2D-canvas skills, creating minor overlap risk with closely related skills.

4 / 5

Total

18

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

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.