Deuz-SDK
MCP ServerDeuz-AI/Deuz-SDK
Deuz-SDK is a zero-dependency TypeScript runtime designed for production-grade AI agents, offering advanced durability, state compaction, and memory management.
496 12MITUpdated 2026-08-13
90
ExcellentOverview
Deuz-SDK is an edge-safe, zero-runtime-dependency TypeScript SDK for building resilient AI agents. It addresses complex production challenges such as long-term memory, context window compaction, human-in-the-loop approvals, resumable execution checkpoints, and Model Context Protocol (MCP) integrations across dozens of model providers.
Capabilities
- ▸Zero runtime dependencies and edge-safe design
- ▸Advanced conversation compaction and self-healing tool loops
- ▸Durable cross-session memory pipeline with fact extraction and scoring
- ▸Human-in-the-loop validation with HMAC-signed expiring tokens
- ▸Resumable execution via database checkpoints
- ▸Multi-provider chat, embedding, and tool execution support across 28+ providers
- ▸Built-in guardrails for input and output verification
Best for
Building autonomous agent loops with tool execution, planning, and verification, Managing durable, cross-session memory using vector stores, databases, or local files, Running multi-step LLM tasks with automatic message compaction to avoid context limits, Integrating secure Model Context Protocol (MCP) servers with built-in OAuth support, Migrating existing projects from the Vercel AI SDK to a edge-safe runtime
Works with
Claude CodeCursorGitHub CopilotChatGPTCodexGemini CLIOpenCode
Evoa Score breakdown= Σ (score × weight)
90Task 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%95 → +14.3
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%58 → +7.0
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%68 → +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 total90.2 / 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
Content capture is opt-in with strict redaction. API keys and sensitive tokens are prevented from leaking into logs, spans, or error payloads.
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.