VALG-ML-Theory-Agent
DechenZhang/VALG-ML-Theory-Agent
Reviews branch-local global proof diagnostics to evaluate theorem viability and determine if they should be accepted, rewritten, or rejected before step-level proof work.
29 2MITUpdated 2026-08-14
82
GreatOverview
The Global Proof Review skill acts as an early theorem-level diagnostic gate in automated mathematical and theoretical ML verification workflows. It independently audits proof diagnostics against setting files and proof sketches, enforces strict gates regarding mechanism witnesses, noncircular closure, and baseline invariance, and routes flawed ideas to appropriate retry targets.
Capabilities
- ▸Theorem-level diagnostic auditing
- ▸Mechanism witness gate verification
- ▸Noncircular closure and invariance checking
- ▸Early idea-failure and obstruction screening
- ▸Structured status routing (ACCEPTED, REVISE_GLOBAL, REVISE_SKETCH, IDEA_FAIL)
Best for
Evaluating global proof diagnostics prior to executing step-level proof work, Screening mathematical and theoretical machine learning ideas for fundamental obstructions early in the workflow, Routing unviable proof strategies or inadequate global proofs to the correct revision stage
Works with
Claude CodeCursorCodexChatGPTGemini CLIGitHub CopilotOpenCode
Evoa Score breakdown= Σ (score × weight)
82Task usefulness20%88 → +17.6
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%33 → +4.0
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%100 → +15.0
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%38 → +3.0
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 total81.6 / 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
Operates strictly on local markdown proof artifacts and does not execute untrusted external code or network operations.
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.