dreamina-design

Automatically turn a validated DCC preview into an approved Seedance video through Dreamina Design MCP, with submit-once recovery and verified download.

Use when an agent needs to install, update, authenticate, troubleshoot, or operate Dreamina（即梦） CLI across account, session, image, video, task-query, and download workflows.

Use when the user provides local images and wants Dreamina 即梦 image-to-image editing or image upscaling through `dreamina image2image` or `dreamina image_upscale`, including validation, submission, polling, and troubleshooting.

Use when the user provides image, video, or audio references and wants Dreamina 即梦 video generation through `image2video`, `frames2video`, `multiframe2video`, or `multimodal2video`. Covers CLI v1.4.18 runtime ratio discovery, required video resolution, Seedance model constraints, input limits, and async terminal statuses.

Use when the user wants to submit, poll, or troubleshoot Dreamina 即梦 text-to-image tasks through `dreamina text2image`. Covers CLI v1.4.18 required resolution, custom width/height, Seedream model tokens, batch count, sessions, and async terminal status handling.

Use when the user wants to submit, poll, or troubleshoot Dreamina 即梦 text-to-video tasks through `dreamina text2video`. Covers CLI v1.4.18 runtime ratio discovery, required video resolution, Seedance model constraints, sessions, and async terminal statuses.

Dreamina Design invocation spec for WorkBuddy - the image/video generation CLI (trusted_cli), text2image/image2image/text2video/image2video families across cli/opencli/prompt variants, shot annotation, video evaluation, and the paid submit-once gates with real submit_id verification. Read this before any Dreamina generation task.

Thin router Skill for Dreamina Design image and video workflows. Selects between the 13 packaged Dreamina Skills (dreamina-cli-* and dreamina-prompt-*) and never duplicates their prompt or CLI bodies.

Guide Jimeng image-to-image for standard members without dreamina CLI. opencli jimeng has no image2image subcommand; only generate, history, new, and workspaces are allowed. Use with dreamina-prompt-image2image for edit prompts, manual reference upload on the Jimeng web UI in the same browser session, and opencli history for verification. Never invoke dreamina or jimeng-cli execution skills.

Guide Jimeng image-to-video for standard members without dreamina CLI. opencli jimeng has no image2video, frames2video, or multimodal subcommands. Use with dreamina-prompt-image2video for incremental motion prompts, manual mode selection on the Jimeng web UI, and opencli history for verification. Never invoke dreamina or jimeng-cli execution skills.

Execute Jimeng text-to-image for standard (non-VIP) members using only opencli jimeng browser commands on jimeng.jianying.com. Covers generate, history, new, workspaces, user_credit, user_assets, user_subscription. Does not use dreamina CLI. Pair with dreamina-prompt-text2image for prompts. Use when the user cannot use dreamina and must automate via opencli Browser Bridge.

Guide Jimeng text-to-video for standard members without dreamina CLI. Use opencli jimeng generate-video (type=video), history, new, workspaces, and account commands. Use with dreamina-prompt-text2video for motion prompts. Never invoke dreamina or jimeng-cli execution skills.

Provides comprehensive guidance for crafting image-to-image (图生图) edit prompts for 即梦 (Dreamina/Jimeng) models 4.0+. Uses a Keep/Change framework: explicitly declare what stays and what transforms. Use when the user wants to modify an existing image via style transfer (风格迁移), background replacement (背景替换), detail enhancement (细节增强), color adjustment (色彩调整), element add/remove (元素添加移除), season/time change (季节时间变换), restoration (修复), outfit change (换装), creative transformation (创意变形), or multi-reference compositing (多图合成); mentions "图生图", "image2image", "编辑图片", "换背景", "改风格", "修图", "把这张图变成"; or provides a reference image and asks to transform it. Covers 10 edit categories with 30 annotated examples and I2I-specific Keep/Change templates. Always use this skill when the user wants to edit an existing image using 即梦.

Provides comprehensive guidance for crafting image-to-video (图生视频) prompts for 即梦 (Dreamina/Jimeng) Seedance 2.0. Image-to-video prompts are incremental — the reference image provides the visual content, the prompt only describes what moves and changes. This skill covers 4 sub-modes: single image→video (单图生视频), first-last frame transition (首尾帧), multi-frame storyboard narrative (多帧故事), and multi-modal reference compositing (全能参考). Use when the user has reference image(s) and wants to animate them into a video; mentions "图生视频", "image2video", "图片转视频", "让这张图动起来", "首尾帧", "故事板", "多帧故事", "frames2video"; or provides images and asks to generate a video from them. This skill covers 4 sub-mode templates with 18 annotated examples, a mode decision tree, and I2V-specific incremental description vocabulary. Always use this skill when the user wants to animate reference images into video using 即梦.

Provides comprehensive guidance for crafting text-to-image prompts for 即梦 (Dreamina/Jimeng) AI image generation. Use when the user wants to write, refine, or optimize an image generation prompt; mentions "提示词", "文生图", "text2image", "文字成图", "AI绘画", "AI生图", "prompt", "帮我写一个...的画面", "生成一张...的图片", "写一个...的提示词"; describes an image they want to create; or provides a rough image description to be turned into a polished production-ready prompt. This skill covers 35 scenario categories with 115+ annotated examples, a comprehensive word library spanning 60+ thematic subcategories, and complete color references including 159 Chinese traditional colors, the 384-color Forbidden City palette, and international standards (Pantone, RAL). Always use this skill when the user needs help writing, refining, or translating any image prompt for 即梦.

Provides comprehensive guidance for crafting text-to-video prompts for 即梦 (Dreamina/Jimeng) video models (视频3.0 Pro, Doubao Seedance 2.0, Seedance 1.5/1.0, 智能多帧). Video prompts require two additional dimensions beyond image prompts: explicit motion description and camera movement direction. Use when the user wants to write, refine, or optimize a video generation prompt; mentions 文生视频, text2video, 视频提示词, 运镜, camera movement, 生成一段...的视频, 帮我写个视频; describes a scene they want as a video; or provides a rough scene description to be turned into a polished video prompt. This skill covers 12 video scenario categories with 35 annotated examples, a motion and camera word library, and a complete camera movement reference covering 7 basic plus 6 advanced compound techniques with emotional effect mappings. Always use this skill when the user needs help writing any video generation prompt for 即梦.

Resume a paused Dreamina 3D job from its ledger state. Use when the user wants to continue an in-flight job without re-paying or re-exporting.

Use when labelling contact sheets or keyframes with shot semantics before a Dreamina redesign, producing shot_annotation.schema.json content only.

Use when judging generated Dreamina shot clips against measured gates, producing shot_evaluation.schema.json content only.

Use when creating a complete Dreamina video project from a local reference video, including analysis, redesign or authorized replication, batch generation, evaluation, audio, subtitles, composition, recovery, and verified export.

Compare a candidate image against a target image using a fresh-context vision subagent and a 4-dimension rubric (composition / lighting / materials / details, total /10). Use when the user asks for visual evaluation against a target image, wants to score image fidelity, or wants a structured gap list to drive the next iteration of an image-generation prompt; mentions "视觉评审", "视觉对比", "视觉打分", "对标", "目标图对比", "target.png 对比", "看看哪里不像", "和目标图差距", "vision judge", "vision evaluation", "image rubric", "compare candidate to target", "score image fidelity". Sits next to `dreamina-prompt-text2image` / `-image2image` / `-text2video` / `-image2video` — same prompt layer, different output. NEVER confuse with `dreamina-video-evaluator` (the latter judges measured video clip gates against design intent for the Dreamina video project; this skill judges two still images against each other).

Obtain the target image for a visual-quality loop when the user has not supplied one — generate it with the approved paid image tool, or baseline-refine from an existing artifact so the target converges instead of diverging. Use when the user asks to "iterate to a target", has no reference image, says "帮我定个目标图", "生成目标图", "对标图", "target image", "goal image", or when a visual loop needs a target but `lock_target` has nothing to lock. Enforces the anti-concept-art discipline (the target is a screenshot-exact rendering to compare against, NOT a stylized interpretation) and fail-closed behavior without paid-generation approval. NEVER generates without the user's explicit approval — target generation is a paid submission. Do not use for judging (that is `dreamina-vision-judge`).