compact-plus
u-ichi/compact-plus
An intelligent plugin for Claude Code and Codex that preserves and restores working state around context compaction events.
189 5MITUpdated 2026-07-27
78
GreatOverview
The compact-plus plugin transparently enhances conversation compaction by backing up transcripts, generating structured 10-section state summaries via LLM, and reinjecting vital session context post-compaction. It monitors context thresholds to warn users, recites active plans before compaction, and provides a manual fallback skill to prevent context drift and memory loss in long-running agent sessions.
Capabilities
- ▸Pre-compaction transcript backup and 10-section state generation
- ▸Post-compaction state recovery and context reinjection
- ▸Context usage threshold monitoring with prompt reminders
- ▸Active plan and decision recitation before compaction
- ▸Manual state capture via the /compact-plus skill
Best for
Preventing context loss and working memory drift during long AI-assisted software development sessions., Automatically summarizing and restoring critical architectural decisions, active plans, and project state when context compaction triggers., Providing pre-compaction warnings and contextual recitations to keep agents focused before trimming history.
Works with
Claude CodeCodex
Evoa Score breakdown= Σ (score × weight)
78Task 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%36 → +4.3
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%58 → +4.6
コミュニティの採用度。 — GitHub Stars/Forks（対数スケール）
Compatibility5%50 → +2.5
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 total78.1 / 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 shell scripts and background hooks during session life cycle events. It relies on local temporary directories for state storage and invokes secondary LLM CLI processes, requiring trust in the local environment and configured backend commands.
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.