You've heard of prompt engineering — but clever wording only gets you so far. Context engineering is the next step: instead of refining what you ask an AI, you plan the working environment you hand it — the perspective it writes from, the job you are actually asking for, the boundaries it follows, and the source material it is told to use.

In this workshop you'll move beyond one-line prompts and build a reusable Context Brief, a four-part planning scaffold that makes your requirements and your source boundary visible. Just as importantly, you'll practice the step that comes after: checking the output against the source before you use it.

No technical background required. 

A Context Brief is a planning scaffold. It makes requirements and source boundaries explicit, and it makes violations detectable. It does not guarantee that output is accurate, and this workshop does not teach it as a reliability mechanism. A model can misquote, overreach, or contradict a source it was given. Human verification is the step that does that work, and it is built into every brief and every activity here.

By the end of this 60-minute workshop, you will be able to:

• Compare two outputs generated from the same task and source, and identify one source-supported difference attributable to the instructions.
• Organize an AI-assisted task into purpose/task, audience or role, boundaries, and source material.
• Draft a Context Brief containing all four parts plus an explicit verification step.
• Evaluate an output against the brief and the source, identify one unsupported or noncompliant element of the output, and revise the relevant part of the brief.
•  

Registration: Required, in-person.

Workshop Incentive Program: Any student who attends at least 3 library workshops this semester will be eligible to receive a FREE Yale Library tote bag. Learn more.