| name | media-factory |
| description | Generate and deploy multimedia content (SVG infographics, Mermaid diagrams, Pexels photos, Kokoro TTS audio, AI images, AI videos, HeyGen avatars) for learning modules and web platforms. Use when: "Infografik erstellen", "Diagramm", "Bild suchen", "Audio", "Narration", "Podcast", "Illustration", "Video erstellen", "Multimedia", "Media Factory", content needs visuals or audio.
|
| license | MIT |
| version | 1.0 |
| date | "2026-03-12T00:00:00.000Z" |
| agent | DevOps |
| allowed-tools | ["Bash","WebFetch","WebSearch","AskUserQuestion","mcp__MCP_DOCKER__sendMessage"] |
Media Factory
Generate and deploy multimedia content for learning modules and web platforms. Works standalone
or integrated into other skills (4K-Optimizer, Blog-Workflow, Lesson-Creator).
When to Use
- User asks for a visual, diagram, infographic, illustration, photo, or audio
- Another skill detects a multimedia gap (e.g., 4K-Analyzer recommends visuals)
- User says "Media Factory" or any trigger phrase above
Quick Start
- Determine what's needed — read the router decision matrix (router.md)
- Identify the project — ask which platform if not obvious (read branding/variants.json)
- Generate — invoke the appropriate generator sub-file
- Deploy — use deployer.md to send the asset to the target platform
Workflow
Step 1: Understand the Request
Parse what the user needs:
- Explicit tool request? → Use that tool directly (e.g., "Nutze Pexels" → image-search.md)
- Topic + context? → Consult router.md decision matrix for recommendation
- From another skill? → The calling skill provides context and target
Step 2: Select Project Branding
Read claude-skills/media-factory/branding/variants.json to get the color palette, font, and style
for the target project. If the target is unclear, ask the user:
"Für welches Projekt? (dirk-schulenburg / cannabis-kultur / stecklingsmeister / lernmodule / bswi)"
Step 3: Generate
Invoke the appropriate generator file:
| Generator | File | Cost |
|---|
| SVG Infographic/Diagram | generators/svg-generator.md | Free |
| Mermaid Flowchart | generators/mermaid-generator.md | Free |
| Pexels Stock Photo | generators/image-search.md | Free |
| Illustration (Undraw Library) | generators/illustration-search.md | Free |
| Kokoro TTS Audio | generators/tts-audio.md | Free |
| AI Image (Nano Banana) | generators/ai-image.md | ~$0.04-0.24 |
| AI Video (Veo 3.1) | generators/ai-video.md | ~$0.15-0.75/s |
| Avatar Video (HeyGen) | generators/avatar-video.md | ~$1/min |
| Canva Design | generators/canva-design.md | Free (Edu) |
Step 4: Deploy
Read deployer.md and upload the generated asset to the target platform.
Step 5: Log
For paid generators, append a line to _assets/media-factory/cost-log.jsonl.
Cost Control
Free generators (SVG, Mermaid, Pexels, Illustration Library, Kokoro TTS): Run automatically.
Paid generators (Nano Banana, Veo, HeyGen): ALWAYS show estimated cost and ask for confirmation:
Ich würde ein [TOOL] [ASSET-TYPE] generieren:
Prompt: "[PROMPT]"
Geschätzte Kosten: ~$X.XX
Generieren? (j/n)
Read cost-log on startup: cat _assets/media-factory/cost-log.jsonl | ... to show month-to-date.
Integration
Other skills can invoke Media Factory by including this instruction:
"Read and follow claude-skills/media-factory/SKILL.md to generate [TYPE] for [TOPIC]."
