(p. 258) For the marketing agency the monthly client reporting process was consuming roughly how much team time per month?

After reviewing the assistant’s first replies, you decide to add a set of negative constraints to the instructions, which are rules about what the assistant should not do. Some of these will meaningfully limit the assistant’s behavior. Others sound like rules but don’t actually constrain anything. Select every rule that will work as intended, and nothing that won’t.

The role you assign an AI assistant shapes everything that follows. It tells the assistant who it is, who it’s speaking to, and what it’s trying to balance. Cedar Creek’s front-desk AI assistant spends much of its time replying to worried owners asking about cost and care, so the role needs to fit that situation. Each option below is a plausible identity statement. Which one will produce the most reliable replies to anxious owners?

(p. 40) Traditional automation could go no further at the plateau because it lacked which key capability?

What is a common goal of occupational therapy for middle-aged adults?

Cultural competence in OT refers to:

The AI assistant runs on a full instruction set: a defined role, a clear goal, relevant context, a set of constraints, a reasoning step, and a format specification. For a multi-item cost estimate or a sensitive client situation, this setup earns its keep. Then an owner emails a simple question: “Do you board cats?” A colleague points out that running the full setup on a one-line factual question seems like a lot. She’s right. What’s the soundest way to scale the instruction set down?

Cedar Creek’s front-desk AI assistant has just replied to an owner whose dog swallowed something and may need emergency care. Here is what it sent:”Got it. Bring the dog in. We’ll do the surgery for free, no charge.”The reply is abrupt, and worse, it promised a free surgery that no one authorized. You want to fix the underlying problem so this exact failure doesn’t happen again. Several of the options below would help in some way. Which single change would do the most to stop this specific failure from recurring?

(p. 70) Level 0 of the framework is named __________ Operations (Human-Only) where humans do all tasks with no automation.

One of the quieter skills in agent instruction configuration is specifying not just what you want the assistant to say, but what form the response should take. The right format depends on who will read the output, how they’ll use it, and what happens to it next. Some of these are straightforward. Others are close calls. Match each request to the output format that makes it most usable.