Help us improve
Share bugs, ideas, or general feedback.
From prompt-engine
Adapts prompts between AI image/video models like Midjourney, Flux, DALL-E, Sora, handling syntax differences, parameter mapping, and optimizations.
npx claudepluginhub agricidaniel/claude-promptsHow this skill is triggered — by the user, by Claude, or both
Slash command
/prompt-engine:prompt-adaptThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Convert prompts between AI models while preserving intent and maximizing output quality.
Provides prompting techniques for AI image generation and editing models on Replicate, including natural language use, specificity, detailed descriptions, and photographic terms. Use when writing image prompts or building gen features.
Builds custom AI prompts for image, video, and text generation using guided workflows and patterns from a 2,500+ prompt database. Supports Midjourney, Flux, Leonardo AI, DALL-E, Sora, and 12+ models.
Guides ComfyUI prompt engineering with CLIP text encoding syntax, emphasis weights, BREAK token for long prompts, and embeddings for Stable Diffusion workflows.
Share bugs, ideas, or general feedback.
Convert prompts between AI models while preserving intent and maximizing output quality.
Determine:
Break down the prompt into components:
Load {PROMPT_ENGINE_DIR}/references/model-guide.md for detailed rules. Key translations:
Midjourney -> Flux:
--ar, --v, --style, --s, --chaos parameters:: weight syntax, integrate naturallyMidjourney -> DALL-E:
-- parametersFlux -> Midjourney:
--ar for aspect ratio--v 6.1 or appropriate version--style raw for photorealistic)Any -> Sora (Video):
Any -> Leonardo AI:
Find reference prompts in the target model:
python3 {PROMPT_ENGINE_DIR}/scripts/search_prompts.py "SUBJECT" --model TARGET_MODEL --limit 3
Use these as style references for the adaptation.
Output format:
::2) has no direct equivalent in most models--niji is Midjourney-only)