kameo

What This Skill Does

The Kameo AI skill allows you to transform static portraits or avatar images into realistic, 5-second talking-head videos. By leveraging advanced facial animation, lip-syncing, and movement generation, this tool breathes life into still assets. It is designed to handle everything from simple dialogue to complex, context-aware emotional performances, making it an essential utility for content creators, marketers, and developers looking to add a personalized touch to their digital communications.

Installation

You can integrate this skill into your environment using the OpenClaw CLI:

clawhub install openclaw/skills/skills/veya2ztn/kameo

Once installed, ensure your Kameo API key is configured. You can either set the KAMEO_API_KEY environment variable or create a configuration file at ~/.config/kameo/credentials.json. For first-time users, we provide a helper script at scripts/register.sh to guide you through the authentication process.

Use Cases

• Personalized Outreach: Create custom video messages for email campaigns or social media outreach.
• Virtual Assistants: Animate static AI-generated character portraits to act as face-of-brand agents.
• Content Localization: Sync dialogue in various languages with a consistent avatar.
• Educational Narratives: Bring historical figures or characters to life for dynamic, engaging presentations.
• Storytelling: Add dramatic flair to static visual media without the need for high-end studio equipment.

Example Prompts

1. "Take my profile photo and make it say 'Welcome to the team! I am so excited to start our project together' in a professional and friendly tone."
2. "Animate this portrait of a sci-fi character. Describe the futuristic lab background, the person's clinical look, and have them say, 'System integration is complete.'"
3. "Create a 5-second video from this image of a woman in a park. She should look surprised and say, 'I can't believe it actually worked!' while maintaining a cheerful expression."

Tips & Limitations

To achieve the highest quality results, treat the prompt as a cinematic instruction rather than just a script. Include details about lighting, camera angle, and the character's expression to help the AI model align the movement with the image's aesthetic.

Best Practices:

• Always use high-resolution, clear portraits where the eyes and mouth are visible.
• Utilize a vision model to pre-analyze your image so the prompt reflects the existing setting.
• Note that the generation process takes roughly 10-30 seconds. Plan your workflow to account for this latency.
• Keep in mind that the current limit is 5 seconds per generation; keep your dialogue concise to ensure the lip-sync remains accurate and natural throughout the entire duration.