Decision Trees

**Judgement:** This is a strong general-purpose skill for structured decision-making whenever outcomes and rough probabilities can be quantified. It excels at turning fuzzy choices into explicit decision trees with expected value calculations and transparent recommendations. **Key strengths & risks:** - **Strengths:** Clear workflow, good cross-domain examples (business, trading, personal), and emphasis on visual/tree structure and EV maths. It repeatedly highlights limitations and encourages users to think through all branches and worst cases, which improves decision quality. - **Risks:** Assumes the user (or assistant) can provide meaningful probability and value estimates; if these are overly subjective or biased, the outputs can create a false sense of precision. It also assumes risk neutrality (EV focus) and does not deeply model risk aversion, emotional, ethical, or true-uncertainty (“black swan”) scenarios. **Recommended scenarios:** Use this skill when the user - has multiple options with **measurable consequences** (money, utility, points, etc.), - can at least roughly estimate **probabilities** of key outcomes, and - wants a **transparent, EV-based comparison** of alternatives (business strategy, investments, capacity planning, hiring, career moves, major purchases). Avoid relying on it for decisions dominated by **deep uncertainty, purely emotional/ethical tradeoffs, or high-stakes situations where probability estimates are very speculative**; in those cases, present it as a thinking aid rather than a definitive optimizer.