メニュー
Use when building AI-powered products or agents, when raw model intelligence isn't enough to solve user problems, or when designing the architecture for agentic workflows
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when designing AI agent products, defining roadmaps for agentic workflows, or evaluating how to evolve AI from passive tool to proactive partner in software development
Shift from rigid roles (PM, Designer, Engineer) towards a model where individuals use AI to bridge skill gaps, enabling smaller and flatter teams. Use in early-stage startups or innovation units where speed is critical.
Use when designing AI agents or chatbots, when raw model outputs feel like a black box, or when building natural language interfaces that need structure and trust
View yourself as an entity with infinite dimensions rather than a single good/bad identity. Strengths and weaknesses are often the same trait in different contexts. Use during performance reviews or when receiving tough feedback.
Review your calendar, color-code activities by energy impact (Green/Red), and systematically delegate draining tasks to maximize time in your Zone of Genius.
Use when dealing with ecosystems, network effects, or high-uncertainty environments where the right answer cannot be known in advance, when rigid planning consumes more value than it creates
| name | three-layer-agent-stack |
| description | Use when building AI-powered products or agents, when raw model intelligence isn't enough to solve user problems, or when designing the architecture for agentic workflows |
A framework for building effective AI agents by synchronizing innovation across three distinct layers: Model, API, and Harness. Success requires tight integration—not treating the model as a black box.
Core principle: Features like "compaction" (long-running tasks) require simultaneous changes across all three layers.
┌─────────────────────────────────────────────────────────────────┐
│ LAYER 3: HARNESS / PRODUCT LAYER │
│ ───────────────────────────────────────────────────────────── │
│ The environment that executes actions and provides context │
│ • VS Code / IDE integration │
│ • Terminal / Shell access │
│ • Sandbox / Secure execution environment │
├─────────────────────────────────────────────────────────────────┤
│ LAYER 2: API LAYER │
│ ───────────────────────────────────────────────────────────── │
│ Interface handling state, context windows, and orchestration │
│ • Context management / Compaction │
│ • State handoff between sessions │
│ • Tool routing and formatting │
├─────────────────────────────────────────────────────────────────┤
│ LAYER 1: MODEL LAYER │
│ ───────────────────────────────────────────────────────────── │
│ Foundation model providing reasoning and intelligence │
│ • Code generation / Reasoning │
│ • Summarization for compaction │
│ • Environment-specific training │
└─────────────────────────────────────────────────────────────────┘
| Principle | Description |
|---|---|
| Full-Stack Iteration | Changes often need Model + API + Harness together |
| Harness Specificity | Models perform best when trained for specific environments |
| Feedback Loops | Product usage (Harness) must inform model training |
| Safety Sandboxing | Harness provides secure environment for code execution |
Implementing "Compaction" to allow Codex to run 24 hours:
Source: Alexander Embiricos (OpenAI Codex) via Lenny's Podcast