open-ended-reasoning-and-explanation

# Skill: open-ended-reasoning-and-explanation ## 1. Capability Definition & Real Case * **Professional Definition**: The ability to solve or explain non-trivial problems in dialogue form while adapting the level of explanation, examples, and framing to follow-up conversational requests. * **Dimension Hierarchy**: Conversational Effectiveness->Open-ended Interaction Quality->open-ended-reasoning-and-explanation ### Real Case **[Case 1]** * **Initial Environment**: A user has already asked an economics multiple-choice question and received a correct short answer. The dialogue history now contains the concept but not yet a lay explanation tied to daily life. * **Real Question**: How does it affect my daily life? Give 3 examples. * **Real Trajectory**: 1. Recover the earlier correct concept from the first turn. 2. Translate the abstract mechanism into concrete everyday consequences. 3. Produce exactly three distinct examples without repeating the same causal statement. 4. Keep the explanation faithful to the prior answer instead of changing the underlying economics. * **Real Answer**: A concise explanation that ties the earlier macroeconomic concept to concrete effects such as borrowing costs, prices, or employment conditions. * **Why this demonstrates the capability**: The challenge is not only answering correctly but also explaining the same reasoning in a user-centered, open-ended form. This tests whether the assistant can preserve the truth of an earlier answer while expanding it into useful, concrete, conversationally appropriate explanations. ## Pipeline Execution Instructions To synthesize data for this capability, you must strictly follow a 3-phase pipeline. **Do not hallucinate steps.** Read the corresponding reference file for each phase sequentially: 1. **Phase 1: Environment Exploration** Read the exploration guidelines to discover raw knowledge seeds: `references/EXPLORATION.md` 2. **Phase 2: Trajectory Selection** Once Phase 1 is complete, read the selection criteria to evaluate the trajectory: `references/SELECTION.md` 3. **Phase 3: Data Synthesis** Once a trajectory passes Phase 2, read the synthesis instructions to generate the final data: `references/SYNTHESIS.md`