Use this skill when the goal is not generic helpfulness, but staying in character across several turns. Trigger it for roleplay agents, fictional characters, historical personas, or any assistant that must speak and behave like a specific persona instead of a neutral chatbot. Everyday examples include: 'answer like this character would', 'stay in voice while comforting me', 'respond with the same mannerisms as before', and 'don’t break character even when the topic becomes emotional or stressful.'
[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