Use when the user wants to convert a chosen visual concept into an optimised text-to-image, image-to-image, or video prompt.
npx claudepluginhub danielrosehill/claude-code-plugins --plugin visual-communicationsThis skill uses the workspace's default tool permissions.
Transform a visual concept into a detailed, model-optimized prompt for image or video generation with 2–3 variations and model-specific guidance.
Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Share bugs, ideas, or general feedback.
Transform a visual concept into a detailed, model-optimized prompt for image or video generation with 2–3 variations and model-specific guidance.
Load project context. Check for an active project; if not, ask the user to specify. Read project.yaml for style preferences, target platform, and resolution guidelines.
Clarify the visual concept. If the user references an idea from ideation.md, read that file. If providing a new concept, confirm the visual type, mood, and purpose.
Determine generation method. Identify whether this is text-to-image, image-to-image, or video generation based on the concept and user intent.
Structure the prompt. Use this framework: [Subject] + [Style] + [Composition] + [Lighting/Mood] + [Technical Specs]
Generate 2–3 prompt variations. Create variations that:
Add negative prompts. If relevant, include negative prompts to avoid unwanted elements (blurry, low quality, watermarks, oversaturated, stock photo feel, etc.).
Provide model-specific tips. Recommend:
Save the prompt YAML. Write to ${CLAUDE_PLUGIN_DATA:-$HOME/.local/share/claude-plugins}/visual-communications/projects/<project-name>/prompts/<slug>.yaml:
name: <Visual Name>
created: <date>
model: <flux | sdxl | video>
prompt: |
<full generation prompt>
negative_prompt: |
<optional negative prompt>
settings:
aspect_ratio: <ratio>
style: <style>
resolution: <resolution>
variations:
- prompt: |
<variation 2>
- prompt: |
<variation 3>
generated_files: []
notes: |
<any additional guidance or iteration notes>
Present to user. Show all prompts with recommended settings. Solicit feedback, iterate if needed, and finalize.
projects/<project-name>/prompts/<slug>.yamlgenerated_files list initialized (to be populated after generation)/run-fal-generation to execute