#
ENJA

marks-pi-harness

earlyaidopters/marks-pi-harness

A comprehensive configuration harness for running local coding agents on Apple Silicon without cloud dependencies or API bills.

7 6MITUpdated 2026-08-10

Overview

This repository provides a production-grade configuration harness for the pi coding agent running locally via LM Studio on macOS. It includes standing operational rules, custom TypeScript extensions, browser verification loops, and specialized skills designed to make local models perform reliable end-to-end tasks with zero tool errors.

Capabilities

  • Headless browser inspection and visual verification
  • Background process management for long-running servers
  • Offline-aware package management and tool gating
  • Structured session continuity and skill routing

Best for

Running fully autonomous local coding agent sessions without cloud APIs, Executing local web development workflows with visual browser verification, Conducting structured deep research within smaller context windows, Managing background servers and offline environments reliably

Works with

Codex
Evoa Score breakdown= Σ (score × weight)
68
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%90 → +13.5

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%39 → +4.7

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%50 → +7.5

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%85 → +8.5

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%33 → +2.6

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

Compatibility5%25 → +1.3

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 total68.5 / 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

Executes local bash commands and modifies local file systems. Designed for secure sandboxed local developer environments.

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.