Skip to book content
United We TransformCreate teamsGrade your agenda
Free and ungated

Four portable UWT AI skills.

Use complete, bounded prompts for clarity diagnosis, Canvas coaching, STACK engagement design, and human-AI workflow architecture. Each skill includes inputs, workflow, outputs, guardrails, and source concepts.

Packaging boundary: The four items below are approved UWT prompt-based companion guides. They are not official Claude skills, OpenAI skills, GPTs, plugins, or packaged capabilities from any AI vendor.

How to use them safely.

Choose one bounded job, supply evidence you are permitted to use, keep an accountable human in the loop, inspect assumptions and sources, and treat the output as a working artifact rather than an autonomous decision.

01

Choose one consequential job

Begin with a decision, engagement, or workflow that has a clear owner and boundary.

02

Give the model real context

Provide evidence, constraints, decision rights, and known unknowns. Do not ask it to fill gaps with confidence.

03

Validate before acting

Review sources, dissent, risk, privacy, and consequences with the people accountable for the work.