Jimeng SD2 Prompting

Bottom-Layer Reasoning

Apply the think-one-step-further mechanism as a lightweight check:

• Confirm this skill is solving the user's real intent, not only the surface request.
• Make the output immediately usable and name any unavoidable next action.
• Extract the reusable structure and, when aligned, propagate it to adjacent prompts, skills, checklists, or workflows.
• Add one guardrail for the most likely next failure while preserving this skill's primary workflow.

Core Intent

Use this skill to turn loose visual ideas into controllable 即梦 SD2 / Seedance 2.0 prompts. Treat the model like a director-level multimodal video system: define references, subject, action, camera, light, rhythm, sound, and constraints.

Default to Chinese prompts for 即梦 UI unless the user asks for English. Keep prompts production-ready, compact, and directly pasteable.

First Decision