marcus
painted-porch/marcus
Epictetus is an AI code auditor and grading system that systematically evaluates software projects, code quality, and the agents or developers who built them.
14 11MITUpdated 2026-08-10
80
GreatOverview
Acting as a methodical and evidence-based Senior Software Architect, Epictetus performs multi-phase codebase audits against standardized rubrics. It assesses architecture, correctness, testing, ghost code, and authorship cohesiveness, optionally integrating tmux session logs and API timelines from multi-agent experiments to grade output and track development quality over time.
Capabilities
- ▸Project structure and git history reconnaissance
- ▸Systematic line-by-line code review with file and line references
- ▸Specification versus implementation comparison
- ▸Automated smoke testing and test suite evaluation
- ▸Process evidence collection from tmux logs
- ▸Agent interrogation and scoring across multiple dimensions
- ▸Structured JSON and Markdown report generation
Best for
Auditing code quality and correctness of software projects, Grading multi-agent AI coding experiments, Evaluating test coverage and smoke testing application endpoints, Analyzing authorship cohesiveness and contribution distribution
Works with
Claude CodeCodexChatGPTGemini CLICursorGitHub CopilotOpenCode
Evoa Score breakdown= Σ (score × weight)
80Task usefulness20%100 → +20.0
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%88 → +13.2
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%26 → +3.1
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%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%39 → +3.1
コミュニティの採用度。 — GitHub Stars/Forks（対数スケール）
Compatibility5%100 → +5.0
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 total79.9 / 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 shell commands, git queries, and tmux log captures which could potentially expose sensitive repository details if run in untrusted 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.