Building reusable prompt libraries: from snippets to shared templates — companion videos

Tina Huang

Tina presents a five-part task/context/references/evaluate/iterate framework and four named iteration moves. Those elements can help you draft a candidate schema, but the agent and chain-of-thought examples reflect the course's time and context. Test any extracted pattern on representative cases before adding it to an approved library.

What you should get from this: Extract a candidate prompt structure, then add ownership, versioning, evaluation, limits and fallback fields before review.

Watch or know first: Comfortable prompting on your own; this is about turning that into shared assets.

Lawton Learns

The video demonstrates a C.R.A.F.T. meta-prompt (Context, Role, Action, Format, Target audience). It is a useful candidate for testing how a prompt-generating prompt behaves, but its outputs are not guaranteed to be consistent, correct or suitable for a team's use cases.

What you should get from this: Evaluate C.R.A.F.T. as one candidate structure for generating prompts, using representative cases and explicit acceptance criteria.

Watch or know first: None once you know basic prompt structure.