CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

avatar-rendering

Live2D / VRM / MMD 形象渲染这一层的不变式和踩过的坑:空闲降帧与 Pet 窗口帧率总闸、Live2D SDK 能改到哪一层、VRM 加载竞态、形象周边的弹窗和点击穿透、内置模型与动作文件处理。改 static/live2d、static/vrm、static/mmd、static/avatar、static/frame-pacing.js、static/subtitle/subtitle-window.js,或者要讨论「渲染性能还能怎么优化」之前读。

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

93%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.

An exemplary maintenance-knowledge skill body: dense, project-specific invariants with exact identifiers and validation guidance, structured as an overview that defers detail to real, well-indexed one-level-deep reference files. The only structural gap is that guidance is presented as invariants and checkpoints rather than an explicit stepwise workflow with feedback loops.

DimensionReasoningScore

Conciseness

Every bullet carries codebase-specific invariants — exact identifiers (_enterIdleTickMode, window.nekoFramePacing.requestPacedFrame), issue references (#3125), and mechanism-level rationales — with no generic concept tutorials; the opening '已有文档不在这里重复' list actively avoids duplication. Nothing reads as padding; each token is information Claude could not know otherwise.

5 / 5

Actionability

Guidance is fully concrete for an instruction-only skill: exact method names and flags, a ready-to-use regex (/^\/chat(?:_full)?(?:\/|$)/), a copy-paste rAF shim (window.requestAnimationFrame = cb => setTimeout(cb, 16)), named scripts (scripts/smooth_yui_mouth.py), specific tests and CI wiring, and explicit diagnostic ordering ('先量 getComputedStyle(el).pointerEvents,不是先查 z-index').

5 / 5

Workflow Clarity

The macro sequence is coherent and well-ordered — '先分清代码跑在哪个页面' first, then per-domain invariants, then a dedicated '验证时的坑' section with explicit CI/local test validation guidance — but the material is organized as invariants rather than explicit numbered steps with feedback loops, so it sits below the top anchor.

4 / 5

Progressive Disclosure

Each section signals one-level-deep detail ('细节见 references/X.md'), all six referenced files exist in the bundle, no reference file nests further references, and a closing 'references 索引' gives per-file read triggers — a clear overview with well-signaled, appropriately split navigation.

5 / 5

Total

19

/

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 concrete, enumerated capabilities with exact file-path triggers and an explicit read-before clause covering both edit and discussion scenarios. The only weaknesses are a few missing natural symptom phrasings and unaddressed overlap with adjacent VRM/MMD sibling skills.

DimensionReasoningScore

Specificity

The description enumerates multiple concrete capability areas — '空闲降帧与 Pet 窗口帧率总闸', 'Live2D SDK 能改到哪一层', 'VRM 加载竞态', '形象周边的弹窗和点击穿透', '内置模型与动作文件处理' — and backs them with exact file paths (static/live2d, static/vrm, static/mmd, static/avatar, static/frame-pacing.js, static/subtitle/subtitle-window.js), which is comprehensive coverage of the skill's scope rather than a minor-gap listing.

5 / 5

Completeness

Both parts are explicit: 'what' is stated as invariants and pitfalls across the enumerated areas, and 'when' is given with concrete trigger phrases — '改 static/live2d、static/vrm … 或者要讨论「渲染性能还能怎么优化」之前读'. This matches the anchor where both what and when are clearly answered with concrete triggers.

5 / 5

Trigger Term Quality

Good keyword coverage: the technology names (Live2D, VRM, MMD), concept terms (降帧, 弹窗, 点击穿透, 渲染性能), six concrete file paths, and a verbatim natural question 「渲染性能还能怎么优化」. A few natural user phrasings are missing (e.g. 卡顿/掉帧, 换模型/切角色), which keeps it just below the comprehensive-synonym anchor.

4 / 5

Distinctiveness Conflict Risk

The niche is clear (this codebase's avatar-rendering layer) and triggers are scoped by exact file paths, but closely related sibling skills named in the body (vrm-physics, ammo-mmd-physics-hacks, 3d-interaction, vrm-mtoon-outline) occupy overlapping VRM/MMD territory and the description gives no boundary disambiguation — minor overlap risk, matching the anchor below the top level.

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

referenced_paths_exist

Referenced path issues: 2 possibly missing (scripts/slim_live2d_motions.py, scripts/smooth_yui_mouth.py)

Warning

Total

15

/

16

Passed

Repository

Project-N-E-K-O/N.E.K.O

Path

.agent/skills/avatar-rendering/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.