CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

r3f-best-practices

React Three Fiber (R3F) and Poimandres ecosystem best practices. Use when writing, reviewing, or optimizing R3F code. Triggers on tasks involving @react-three/fiber, @react-three/drei, zustand, @react-three/postprocessing, @react-three/rapier, or leva.

Quality

77%

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/r3f-best-practices/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

61%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 well-organized rule index with genuinely useful executable examples for the highest-impact patterns, and good token discipline overall. Its two real defects are the broken reference layer — every referenced `rules/*.md` file is missing from the bundle, so the promised detail tier doesn't exist — and internal inconsistencies ("12 categories" claimed vs. 11 listed; `useStore` vs. `useGameStore`).

Suggestions

Ship the seven referenced rule files (rules/perf-never-set-state-in-useframe.md, rules/perf-zustand-selectors.md, rules/perf-visibility-toggle.md, rules/perf-r3f-perf.md, rules/frame-delta-time.md, rules/drei-use-gltf.md, rules/loading-suspense.md) in the bundle, or remove the dangling references — the "How to Use" section promises files that are not present.

Add a short explicit navigation workflow (task → category prefix → rule file or one-line summary) so the consumption order is unambiguous for a lookup-style skill.

Fix internal inconsistencies: the header claims "70+ rules across 12 categories" while only 11 categories are listed, and the Zustand "BETTER" example switches from `useGameStore` to an undefined `useStore`.

DimensionReasoningScore

Conciseness

The body is mostly lean — one-line rule summaries, code only for critical patterns, and no explanation of basics — but the "Ecosystem Coverage" section spends eight lines describing what each library is ("React renderer for Three.js", "Physics engine"), knowledge Claude already has, and "When to Apply" duplicates the frontmatter triggers. This matches "efficient; minor instances of over-explanation that could be trimmed" rather than the padded anchor 3 case.

4 / 5

Actionability

The critical patterns are shown as concrete, executable BAD/GOOD JSX (setState-in-useFrame vs. ref mutation, Zustand selectors, useGLTF with preload, Suspense, Perf) and the one-liners are terse directives. Minor gaps keep it below anchor 5: the "BETTER" Zustand snippet calls an undefined `useStore` (the earlier example uses `useGameStore`), and `useRef()` is untyped.

4 / 5

Workflow Clarity

This is a reference catalog rather than a procedural skill; the priority table and "How to Use" section give consumption guidance, but the navigation path (identify the task's category → look up its prefix → read the rule file or one-liner) is left implicit and there are no checkpoints or feedback loops. It sits at "steps listed but sequence/checkpoints implicit" rather than anchor 4's clear sequenced procedure.

3 / 5

Progressive Disclosure

The design is correct — one-line summaries in SKILL.md with detail in one-level-deep `rules/*.md` files, clearly listed — but the `rules/` directory does not exist in the bundle: none of the seven referenced files (e.g., `rules/perf-never-set-state-in-useframe.md`) ship, even though the text asserts "Only these rule files ship with this copy". The references dead-end, which is worse than the merely-unclear-signaling anchor 3 and lands at anchor 2.

2 / 5

Total

13

/

20

Passed

Description

82%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: precise domain, explicit and concrete "Use when / Triggers on" guidance, and natural package-name trigger terms with the R3F acronym synonym. Its main weakness is that the "what" rests on the vague phrase "best practices" without naming the practice areas the skill actually covers.

DimensionReasoningScore

Specificity

"Use when writing, reviewing, or optimizing R3F code" lists several action contexts and the domain is pinned by six exact package names, matching the "several specific actions; minor gaps" anchor. Not a 5 because the practice areas covered (performance, state management, physics, etc.) are never enumerated; not a 3 because it goes beyond 1-2 actions with a highly specified domain.

4 / 5

Completeness

The "when" is fully explicit with concrete trigger phrases ("Use when writing, reviewing, or optimizing R3F code. Triggers on tasks involving..."), exceeding the anchor-4 example, but the "what" — "ecosystem best practices" — is thinner than the anchor-5 example's concrete capability list, so it falls between anchors 4 and 5.

4 / 5

Trigger Term Quality

"React Three Fiber (R3F)" pairs the full name with its acronym synonym, and "@react-three/fiber, @react-three/drei, zustand, @react-three/postprocessing, @react-three/rapier, or leva" lists every ecosystem package verbatim — exactly the terms users of this niche naturally say. Omitting bare "three.js" is correct scoping rather than a missing synonym, so this fits the comprehensive-coverage anchor.

5 / 5

Distinctiveness Conflict Risk

A clear niche with precise package-name triggers, but "zustand" and "leva" would fire on non-3D state-management or debug-GUI tasks better served by a dedicated skill, which is exactly the "minor overlap risk with closely related skills" anchor. Not a 5 because of that overlap; not a 3 since the core R3F triggers are unambiguous.

4 / 5

Total

17

/

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

zebbern/claude-code-guide

Path

skills/r3f-best-practices/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.