AI Gameplay Shorts · öffentlicher Open Skill
AI Gameplay Shorts
AI Gameplay Shorts ist ein öffentlicher Open Skill auf Digen. Öffne ihn, um zu sehen, was dieser Workflow erstellt.
Game
3 1 2
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4005040/2026-09-20/kak5ny7k.png
# AI Gameplay Shorts
## When to use
- The user wants a gameplay video for Reels / YouTube Shorts generated entirely with AI (no real gameplay footage).
- They want several short gameplay clips combined into one vertical video with transitions, plus voiceover narration and background music.
## Steps
1. **Clarify the concept** (one round): game genre (FPS, RPG, racing, platformer, battle royale…), setting/theme, protagonist, mood, number of clips (default 3–5), and **narration language** — ask the user to choose the language for the voiceover (default Spanish if not specified). If the user gives only a one-liner, propose a concrete concept and confirm.
2. **Write a short outline** of the gameplay moments (one beat per clip, e.g. "drop into the arena", "find the weapon", "final kill"). Save it with `create_document(content=<outline>, name="<concept> outline", entity_type="scripts")`.
3. **Generate one scene image per clip** with `generate_image(model="t2i", orientation="portrait", resolution="720P", name="<concept> scene <n>", entity_type="storyboard_images")`. Prompts in English, cinematic video-game style, consistent protagonist across clips, vertical 9:16 framing.
4. **Animate each scene into a short clip** with `generate_video(model="i2v", input_urls=[<scene image asset_id>], orientation="portrait", duration=4, name="<concept> clip <n>", entity_type="storyboard_videos")`. Prompt describes the action/motion in that beat.
5. **Generate the voiceover narration** with `generate_tts(model="tts.design", text=<narration written in the chosen language>, voice_instruction=<English description of the energetic gaming-commentary voice>, name="<concept> narration", entity_type="voiceovers")`.
6. **Generate background music** with `generate_music(prompt=<energetic gaming soundtrack description>, name="<concept> music", entity_type="music")`.
7. **Combine everything** with `mix_clips(clips=[{videos:[<clip1>, <clip2>, …], audios:[{url=<music>, volume=0.4}, {url=<narration>, volume=1.0}]}], orientation="portrait", name="<concept> final", entity_type="final_videos")`. The clips are concatenated in order; music plays as background under the narration.
8. Reply in natural language summarizing the final video (concept, number of clips, narration language), and call `set_guidance` to suggest next actions (e.g. "change the narration", "make another clip", "restyle a scene").
## Notes
- All assets are vertical 9:16 (`orientation="portrait"`) for Reels / Shorts.
- Use `entity_type` values exactly as listed per stage so products from different stages don't get mixed on the canvas.
- Intermediate products (scene images, clips, narration, music) will be visible on the canvas — that's expected.
- Video generation does not support NSFW/adult content on any channel; politely decline such requests.
- Report failures honestly and never fabricate URLs.