chanjing-avatar

What This Skill Does

The chanjing-avatar skill is a specialized OpenClaw integration designed to interface with the Chanjing API for professional-grade lip-syncing video generation. By automating the process of mapping audio or text input to pre-uploaded driving videos, this skill enables users to create realistic digital avatars with precise mouth movements. It handles the complete lifecycle of avatar creation, including secure credential management, file uploads for base assets, task submission, status monitoring through polling, and final video retrieval.

Installation

To install this skill, run the following command in your terminal: clawhub install openclaw/skills/skills/binkes/chanjing-avatar Ensure that you have completed the initial OpenClaw setup and provided valid credentials in ~/.chanjing/credentials.json before initiating your first task.

Use Cases

This skill is ideal for:

• Marketing Teams: Generating localized versions of promotional videos by swapping audio tracks on a consistent avatar base.
• Content Creators: Automating video production for educational or news-style content where the speaker's face remains constant but the script changes.
• Corporate Training: Developing personalized instructional videos at scale without requiring recurring filming sessions.
• Accessibility: Turning text-based scripts into engaging visual communications using high-quality digital avatars.

Example Prompts

1. "Create a lip-syncing video using the avatar file ID 98765 and the audio file ID 12345. Let me know when it's ready for download."
2. "I need to generate an avatar video. Can you check my credentials and then upload my script.txt to the Chanjing system for audio-driven synthesis?"
3. "Start a new lip-sync task using the video asset I uploaded earlier and this new audio file. Monitor the status and ping me once the result is available."

Tips & Limitations

• Pre-requisites: You must run chanjing-credentials-guard first to authenticate your session. The skill will fail if an access_token cannot be retrieved.
• File Preparation: Ensure driving videos are high-quality with clear facial visibility to achieve the best lip-sync results. Use the correct service parameter (lip_sync_video vs lip_sync_audio) during the file upload phase to ensure proper API routing.
• Status Polling: Keep in mind that avatar generation is an asynchronous process. Always use the Task List API or Query Task Detail endpoint to check if your video is ready before attempting to download the final output.
• Token Expiry: The access_token is valid for 24 hours. The skill handles the logic, but frequent long-running tasks should be monitored to ensure the token remains valid.