Arnold
peteromallet/Arnold
Arnold and its Megaplan harness provide a structured, cost-efficient pipeline system for coordinating multiple AI models during software development sprints.
Overview
Arnold leverages a native-first pipeline architecture to break complex software development tasks into distinct, independently checked phases like preparation, planning, critique, and execution. By routing specific tasks to cost-effective open models while reserving premium models like Claude or Codex for architectural decisions and hard challenges, it significantly lowers the cost of AI-driven coding without sacrificing reliability.
Capabilities
- ▸Native-first pipeline definition using decorators (@pipeline, @phase, @decision)
- ▸Cost-efficient model routing separating cheap tasks from premium adjudication
- ▸Multi-phase execution harness (Megaplan) for planning, gating, and code execution
- ▸Remote cloud-based pipeline runs with persistent workspaces
Best for
Scaffolding native Arnold pipelines, Running cost-efficient AI coding sprints and epics using model routing, Performing multi-phase software planning, critique, and automated execution
Works with
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
実装・指示の品質。 — AIによるSKILL.md/README解析
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
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
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
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
コミュニティの採用度。 — GitHub Stars/Forks(対数スケール)
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
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
Execution pipelines handle local workspace modifications and interact with external LLM APIs using user-provided credentials stored locally.
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.