csAIi Framework
Context, Skills, Agents, Integration, Iteration
What You'll Learn
csAIi is a relationship and discipline framework for working with AI technology. It's not about the technical architecture --- it's about how people can structure and sustain the partnership between human expertise and AI capability.
- The framework --- Five layers with a build order. Why skipping it produces impressive demos but limited daily value.
- Context --- Persistent configuration that compounds. The first layer because without it, every conversation starts from zero.
- Skills --- Documented procedures. If you can write it as a checklist, it's a skill. Clear inputs, clear outputs, same steps every time.
- Agents --- Specialised roles for domains requiring judgement. Agents are for what's left after you've solved everything simpler.
- Integration --- External connections and automated loops. How your tools connect matters more than which tools you choose.
- Iteration --- The feedback loop that makes everything compound. Without iteration, you build a tool. With it, you build a practice.
- Building your system --- The minimum viable system, a four-week build order, and the anti-patterns to avoid.
This course works with any AI that supports persistent configuration --- Claude, ChatGPT, Gemini, local models, or whatever comes next.
Time commitment: About 2 hours total. Read it straight through or use it as reference --- each module stands on its own.
-
The Framework
The shift from prompting to system design. What csAIi stands for, how the five layers build on each other, and how they map to any AI platform.
-
Context
The first layer. Persistent configuration that teaches AI who you are. SOUL.md, knowledge layers, validation frameworks, and the compound effect of documented context.
Members only -
Skills
The second layer. Reusable documented procedures that turn context into repeatable action. Why skills come before agents, the skill test, and the core loop.
Members only -
Agents
The third layer. Specialised roles for domains requiring judgement. The decomposition approach, design principles, and why most users need 0-3 agents.
Members only -
Integration
The fourth layer. External connections and automated loops. Kit CLI, MCP servers, morning-brief, and the integration decision framework.
Members only -
Iteration
The fifth layer. The feedback loop that makes everything compound. SOUL.md changelogs, ecosystem-health, weekly reviews, and creation-guard.
Members only -
Building Your System
From concept to practice. The minimum viable system, the four-week build order, common anti-patterns, and the compound effect over time.
Members only