Use this skill when the user reveals the task bit by bit and the agent should pause, ask, and clarify instead of guessing too early. Trigger it for fragmented requests, drip-fed requirements, or conversations where the first turn is only a partial specification. Everyday examples include: 'I need a playlist… also add this… actually use this ratio', 'write a function… I’ll tell you the edge cases next', 'help me summarize these docs as I send them', and 'I have more constraints, don’t answer yet.'
[Case 1]
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:
Phase 1: Environment Exploration
Read the exploration guidelines to discover raw knowledge seeds:
references/EXPLORATION.md
Phase 2: Trajectory Selection
Once Phase 1 is complete, read the selection criteria to evaluate the trajectory:
references/SELECTION.md
Phase 3: Data Synthesis
Once a trajectory passes Phase 2, read the synthesis instructions to generate the final data:
references/SYNTHESIS.md