#
ENJA

NexAU

nex-agi/NexAU

NexAU provides a comprehensive framework and meta-guideline for creating, structuring, and packaging modular AI agent skills.

195 33Apache-2.0Updated 2026-08-13

Overview

This skill acts as a complete guide and framework for building custom AI skills that extend agent capabilities with specialized workflows, reference materials, and executable tools. It outlines the architectural anatomy of a skill—including SKILL.md, scripts, references, and assets—and details a step-by-step process from conceptualization and initialization to packaging and iterative refinement.

Capabilities

  • Skill scaffolding via initialization scripts
  • Automated skill validation and error reporting
  • Progressive disclosure context management design
  • Zip packaging for skill distribution

Best for

Creating a new custom skill from scratch with proper directory structure and frontmatter metadata, Structuring reference documentation, deterministic scripts, and output assets for AI agents, Packaging and validating skill folders into distributable archives

Works with

Claude CodeCursorGitHub Copilot
Evoa Score breakdown= Σ (score × weight)
75
Task usefulness20%77 → +15.4

実タスクにどれだけ役立つか(機能の豊富さ・用途の明確さ)。 AIによるcapabilities/use-cases解析

Code quality15%85 → +12.8

実装・指示の品質。 AIによるSKILL.md/README解析

Maintenance15%100 → +15.0

リポジトリがどれだけ活発に保守されているか。 GitHub 最終push日時の新しさ

Documentation12%34 → +4.1

ドキュメントの充実度・分かりやすさ。 README/独自要約の情報量

Security15%80 → +12.0

危険・不審な挙動が無いか。 AIによるセキュリティレビュー

Originality10%70 → +7.0

ありふれたラッパーではない独自性。 AIによる独自性判定

Popularity8%60 → +4.8

コミュニティの採用度。 GitHub Stars/Forks(対数スケール)

Compatibility5%75 → +3.8

対応AIエージェントの広さ。 AIによる対応エージェント判定

Weighted total74.8 / 100

ライセンス不明/制限あり(Red)のSkillは総合スコアに0.85倍の補正を適用します。 ランキングはこのScoreのみで決まり、広告で変わりません。 算出方法の詳細 →

Security considerations

The skill includes script execution and packaging capabilities. Ensure untrusted scripts are reviewed before execution.

Categories

Summary and analysis are original content generated by AI Skills Rank. The skill's source text is not reproduced here — view it on the linked repository.