doubao-capture
Capture Doubao translation results with auto-scroll and auto-end detection.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/banner90/doubao-captureWhat This Skill Does
The doubao-capture skill is a specialized automation tool designed for the OpenClaw AI agent to interface with the Doubao translation interface on Windows. It streamlines the process of extracting long-form translated text by automating scroll actions and identifying when a translation sequence has concluded. By interacting directly with the window handle (HWND) provided by doubao-launch, this agent skill reliably collects subtitle data, manages file output, and handles the synchronization required to turn a live translation session into a structured text file. It is specifically engineered to function within a cross-platform WSL-to-Windows bridge, ensuring that Python scripts executing in Linux can control local Windows GUI applications effectively.
Installation
To install this skill, run the following command in your OpenClaw environment:
clawhub install openclaw/skills/skills/banner90/doubao-capture
Ensure that you have the necessary Windows Python dependencies and the correct driver permissions for GUI automation. The script relies on standard Windows libraries for window management and file system access.
Use Cases
- Content Localization: Automatically transcribing video content by piping audio through Doubao and capturing the resulting localized text.
- Archive Generation: Building a database of translated subtitles from various foreign-language video sources for research or subtitles generation.
- Workflow Automation: Integrating the skill into a batch processing pipeline where audio files are automatically queued, played, captured via Doubao, and saved to the file system without manual intervention.
Example Prompts
- "Execute the translation capture process for the current window handle and save the output to my works folder."
- "Capture subtitles from the active Doubao window and stop automatically when no new lines appear for 10 consecutive reads."
- "Launch the doubao-capture routine with the provided HWND and summarize the final line and character count once done."
Tips & Limitations
- Visibility Requirements: The script requires an active, visible desktop session. You cannot run this skill if you are disconnected from an RDP session, as the GUI automation will fail to detect the Doubao interface elements.
- File Pathing: Ensure the output path is correctly mapped. If using WSL, remember that your Windows directory is accessed via the
/mnt/prefix. - Performance: The
no_new_thresholdparameter can be adjusted based on network latency; increase this value if you notice the capture stops prematurely during slow loading periods.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-banner90-doubao-capture": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read, code-execution
Related Skills
audio-play
Play audio files using Windows media player. Non-blocking execution.
Banner Youtube Translate Workflow
Skill by banner90
doubao-launch
Launch Doubao desktop application and configure real-time translation window.
youtube-audio-download
Download YouTube video audio and convert to MP3. Supports age-restricted videos with cookies.
byt-workflow
YouTube video translation workflow, download audio, launch Doubao, play audio, capture translation