- Ignition: first start: 0/5
- Chassis: car + fuel: 0/4
- Lift and diagnostics: 0/3
- Fuel system: model gateway: 0/5
- Navigator: task route: 0/5 — now
- Glovebox: context and memory: 0/4
- Rain sensor and wipers: pipeline: 0/5
- Hitch and roof rack: MCP, Hooks, Skills: 0/6
- Convoy: subagents and orchestration: 0/6
- Road atlas: AI notebook: 0/5
- Dyno: engine tuning: 0/5
- Onboard computer: second brain: 0/5
You're building: Navigator: task route
Picture a prompt you once fired at AI — and got back something useless.
Now don't think of it as a chat message. Picture it as ordering a bespoke suit from a tailor while mumbling under your breath: "something dark… roughly my size…" The tailor nods politely and cuts blind.
Run it in your head: what kind of "suit" came back? What went unsaid — what did the AI have to guess?
Getting it wrong is fine: it is part of learning. Guessing first helps the answer stick later, and the answer itself comes after the explanation.
Copy your personal prompt and paste it into your agent's learn mode — it picks up your context and takes you forward.
Honestly: with AI it feels like you can do more than you can without it. After practice, check yourself without AI. (Bastani et al., PNAS 2025)
Explains the material using the course method.