- Home
- Skills
- Agents & Orchestration
- agentkit multimedia shopping
agentkit multimedia shopping
Automated multi-modal video generation for Xiao Sheng digital shopping assistants using agentkit-samples.
$5
Works with the AI Agents you already use
agentkit multimedia shopping
Example session with this skill installed
Generate a full 25-second e-commerce video for a business meeting scene. The guide should sound professional and enthusiastic. Follow the Xiao Sheng character profile.
- Read your context and instructions
- Compiled the agentkit multimedia shopping
I have executed the multi-modal workflow to generate the Xiao Sheng digital human video. I generated the character reference image, synthesized a professional voiceover, and composed the background music.
Video Asset Generation Report
| Asset Type | Script Executed | Specifications | Status |
|---|---|---|---|
| Character Image | generate_character.py | 1080x1920 (9:16), Professional Attire | Success |
| Scene Background | generate_scene.py | Business Meeting Room, Neutral Lighting | Success |
| Guide Voice | generate_voice.py | 16kHz Mono WAV, Enthusiastic Tone | Success |
| Background Music | generate_music.py | 16kHz Mono WAV, Piano/Strings | Success |
| Final Video | generate_video.py | 25s Total, 5 Scenes, MP4 | Success |
Next steps
- Review the generated
.wavfiles in the output directory to verify voice tone. - Use the
infinitetalk-shopping-avatarskill to generate specific prompt tags for final InfiniteTalk refinement. - Check the
assets/examples/folder for the final 9:16 MP4 render.
Connects securely to your tools. The creator never sees your data.
About this skill
Manual coordination of image generation, text-to-speech, and video synthesis for e-commerce avatars is slow and prone to visual inconsistencies. This skill automates the production of 9:16 vertical short-form videos featuring the Xiao Sheng shopping guide character. It orchestrates multi-modal workflows to ensure character traits, background aesthetics, and audio formats align with ByteDance agentkit-samples specifications.
What it does
- Character synthesis generates consistent 9:16 vertical images of the Xiao Sheng guide using fixed profiles for face and attire.
- Scene generation creates high-resolution backgrounds for business, office, or study environments with specific lighting profiles.
- Audio production handles TTS voice synthesis and background music generation in 16kHz mono WAV format.
- Video orchestration combines generated assets into a 25-second, 5-scene vertical video using InfiniteTalk logic.
- Workflow automation sequences the transition from character sketches and scene templates to final multi-modal assembly.
How it works
- Character setup invokes
generate_character.pyusing attributes from the character profile reference. - Asset generation triggers scene, voice, and music generation scripts to build the required media library.
- Format conversion ensures all outputs meet the 1080x1920 resolution and specific audio sampling rates.
- Assembly uses the gathered assets to drive the video generation script for the final multi-modal output.
Frameworks & tools
Built for the ByteDance agentkit-samples ecosystem using Python, Pillow, Requests, and NumPy. Integrates with InfiniteTalk for video synthesis and Wav2Vec2 for audio encoding.
Why this beats prompting it yourself
Prompting a standard LLM won't maintain the 9:16 aspect ratio or specific 16kHz mono audio requirements needed for downstream tools. This skill enforces character consistency across multiple generation steps and automates the boilerplate orchestration between separate script files.
Use cases
- Generate professional shopping guide videos for social media marketing.
- Batch produce digital human avatars for e-commerce product demonstrations.
- Create localized marketing content with consistent character branding and background music.
- Prototype vertical short-form video content for live-streaming promotions.
Known limitations
Requires pre-configured API credentials for generation services. Limited to the Xiao Sheng character profile and 9:16 vertical orientation.
Requires pre-configured API credentials. Output restricted to 9:16 vertical format and 25-second durations. Character consistency depends on local reference files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 23 days ago
Permissions
Allowed Hosts