Video To Text

Video to text converter. Downloads videos from Bilibili using bilibili-api, from other sites using yt-dlp, then transcribes audio using faster-whisper. Use w...

Security Scan

The skill's code, instructions, and dependencies align with a legitimate 'video to text' converter; it asks for per-use Bilibili cookies (sensitive) but otherwise contains no unexplained or hidden behavior.

Purpose & Capability

Name/description match the code and README: script downloads videos (bilibili-api or yt-dlp) and transcribes with faster-whisper. Required tools and libraries (yt-dlp, bilibili-api, ffmpeg, faster-whisper) are appropriate for the stated task.

Instruction Scope

SKILL.md and the script limit actions to downloading audio and transcribing. The README instructs users to extract Bilibili cookies via browser DevTools — this is necessary for authenticated Bilibili downloads but is sensitive; the skill does not attempt to transmit credentials anywhere else. The script runs yt-dlp via subprocess and performs HTTP GETs for audio URLs (expected).

Install Mechanism

No automated install spec; instructions tell user to pip install listed packages and ensure ffmpeg. This is low-risk and transparent.

Credentials

The skill does not request environment variables or external credentials by default. It does require Bilibili session cookies for downloading private/age-restricted content; those are provided via CLI or editing the script. Requesting these specific cookies is proportionate to Bilibili access, but storing credentials in the script or exposing them carelessly is a security risk.

Persistence & Privilege

Skill is not always-enabled, does not request special platform privileges, and does not modify other skills or system-wide settings.

Assessment

This skill appears to do what it says: download video/audio and transcribe it locally. Before using it: (1) review the script locally (it’s short and readable) and run in an isolated environment (virtualenv, container). (2) Be cautious when supplying Bilibili credentials: copy SESSDATA/bili_jct/buvid3 only from your browser and do not paste them into shared repos or logs; prefer passing them on the command line for ephemeral use. (3) Ensure yt-dlp and ffmpeg are installed from trusted sources because the script invokes yt-dlp as a subprocess. (4) Expect large model downloads and disk usage for medium/large faster-whisper models. If you want higher assurance, run the script on a machine/account where leaked cookies would have limited impact.

Convert video URLs or local files to text transcripts.

Usage

python3 scripts/video_to_text.py <video_url_or_local_file> [options]

Arguments

Argument	Description	Default
url	Video URL or local file path (required)	-
-m, --model	Whisper model size	base
-l, --language	Specify language code	Auto-detect
-o, --output	Output file path	Print to terminal
--keep-files	Keep downloaded audio/video files	No
--sessdata	Bilibili SESSDATA	From config
--bili-jct	Bilibili bili_jct	From config
--buvid3	Bilibili buvid3	From config

Model Selection

Model	Size	Speed	Accuracy
tiny	~75MB	Fastest	Lowest
base	~150MB	Fast	Basic
small	~500MB	Medium	Good
medium	~1.5GB	Slow	Very Good
large	~3GB	Slowest	Best

Examples

# Bilibili video (requires auth)
python3 scripts/video_to_text.py "https://www.bilibili.com/video/BVxxx"

# Specify Chinese language
python3 scripts/video_to_text.py "https://www.bilibili.com/video/BVxxx" -l zh

# Local file
python3 scripts/video_to_text.py "/path/to/video.mp4" -m small

# Save to file
python3 scripts/video_to_text.py "https://www.bilibili.com/video/BVxxx" -o result.txt

Supported Platforms

• Bilibili (bilibili.com) - Requires auth
• YouTube - via yt-dlp
• TikTok/Douyin - via yt-dlp
• Twitter/X - via yt-dlp
• Any site supported by yt-dlp
• Local files - supports mp4, wav, m4a, webm, mkv, etc.

Bilibili Auth Setup

Method 1: Config File

Edit BILIBILI_CREDENTIALS dict in the script:

BILIBILI_CREDENTIALS = {
    "sessdata": "your_sessdata",
    "bili_jct": "your_bili_jct",
    "buvid3": "your_buvid3"
}

Method 2: Command Line

python3 scripts/video_to_text.py "https://www.bilibili.com/video/BVxxx" \
    --sessdata "xxx" \
    --bili-jct "xxx" \
    --buvid3 "xxx"

How to Get Auth Info

1. Login to Bilibili web (bilibili.com)
2. Press F12 to open Developer Tools
3. Application -> Cookies -> bilibili.com
4. Copy these values:
   • SESSDATA
   • bili_jct
   • buvid3

WARNING: These are your login credentials. Don't share with others!

Installation

# Install dependencies
pip3 install bilibili-api-python yt-dlp faster-whisper aiohttp requests

# Ensure ffmpeg is installed
# Ubuntu/Debian: sudo apt install ffmpeg
# CentOS: sudo yum install ffmpeg

Dependencies

• bilibili-api-python - Bilibili API
• yt-dlp - Video download
• ffmpeg - Audio/video processing
• faster-whisper - Speech transcription
• aiohttp - Async HTTP
• requests - HTTP requests