token-diet
Kulaxyz/token-diet
An always-on token optimization skill for coding agents that reduces session token consumption by eliminating filler and improving context management without sacrificing correctness.
485 4No licenseUpdated 2026-07-04
71
GoodOverview
Token Diet cuts wasted tokens across entire AI agent sessions by enforcing concise communication, lean code artifacts, disciplined tool use, and optimized context search strategies. It operates across multiple intensity levels—including a telegraphic ultra mode—while strictly safeguarding correctness, reasoning depth, and critical test coverage.
Capabilities
- ▸Context window optimization through aggressive grep-first search and batched reads
- ▸Multi-level token reduction (lite, on, ultra modes)
- ▸Elimination of conversational filler, restated prompts, and redundant narratives
- ▸Streamlined artifact creation for documentation, plans, hand-offs, and comments
- ▸Cheaper sub-agent delegation for broad exploratory tasks
Best for
Minimizing API and subscription token costs during prolonged coding and refactoring sessions, Speeding up agent responses by removing conversational preamble, postamble, and verbosity, Enforcing efficient context gathering techniques like targeted grep and batched tool calls instead of reading entire files
Works with
Claude CodeCodexCursorGitHub Copilot
Evoa Score breakdown= Σ (score × weight)
71Task 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%95 → +14.3
実装・指示の品質。 — AIによるSKILL.md/README解析
Maintenance15%85 → +12.8
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%43 → +5.2
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%67 → +5.4
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 total83.6 / 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
The skill instructs agents to minimize unnecessary file reads and maintain security-sensitive verification routines locally, introducing no inherent code execution vulnerabilities.
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.