Subtitle Generator

v1.0.7

当用户需要生成字幕、制作字幕、字幕对齐、ASR识别、语音转文字时使用此技能。触发词:生成字幕, 字幕生成, 制作字幕, 视频字幕, 语音识别, ASR字幕, Whisper字幕, 字幕制作, 视频转字幕, 字幕对齐, 音频转字幕, 语音转文字, 自动字幕, ASR识别, faster-whisper, subti...

0· 334·1 current·1 all-time
bysniper-one@aqbjqtd
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description (subtitle/ASR) align with the included Python code: audio extraction (ffmpeg), ASR engine wrappers (faster-whisper / openai-whisper), timing alignment, and subtitle file generation. No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md instructs the agent to run the bundled CLI script in background and to use openclaw events/message tools for notifications; the code implements that. One small mismatch: SKILL.md claims engines/__init__.py will auto-create the venv on first import, but the code only triggers venv creation when create_engine cannot import an engine (ensure_whisper_venv is called later). Otherwise instructions stay within the subtitle task (no file reads or env access outside the skill's scope).
Install Mechanism
No platform install spec is provided, but the code will auto-create ~/.whisper-venv and run pip install -r requirements.txt (faster-whisper, openai-whisper, ffmpeg-python) when needed. Installing PyPI packages at runtime is expected for this functionality but is a higher-impact action (downloads, possible native builds) than a pure instruction-only skill.
Credentials
The skill requests no environment variables or credentials and only uses Path.home() and standard local commands. The packages installed are proportional to ASR/subtitle generation; no unrelated secret access is requested.
Persistence & Privilege
The skill does not request always:true and does not modify other skills. It will create a per-user venv (~/.whisper-venv) and write subtitle files next to the video—these are reasonable for its purpose.
Assessment
This package appears to be a legitimate local subtitle generator, but be aware it may automatically create ~/.whisper-venv and run pip installs (faster-whisper/openai-whisper/ffmpeg-python) which will download and install code and possibly native wheels. If you want to be cautious: (1) inspect or run the script in an isolated environment or container, (2) ensure you have ffmpeg installed separately, (3) be prepared for potentially long installs or native compilation, and (4) note the skill expects to be run in the background and will use the OpenClaw 'system event' / message flow to notify you (it will not exfiltrate data to remote endpoints). If you need higher assurance, request a signed upstream source/homepage or run the code manually first to review behavior.

Like a lobster shell, security has layers — review code before you run it.

latestvk973st7chqwd9qf58gy7w0pq5183v7t9

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments