VideoTranscribe

Warn

Audited by Snyk on Jun 5, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.85). Outsider free text can be ingested when the user provides a URL: Tools/transcribe.py runs yt-dlp --write-auto-sub/--write-sub ... --skip-download and then reads the downloaded subtitle files (.vtt/.srt) via parse_subtitle_file(), which strips timestamps/HTML and returns the subtitle body text into text that is later sent to the LLM (Gemini) for transcription.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.90). The transcribe.py script installs and imports the google-genai package at runtime (via pip), which fetches and executes remote code from PyPI (e.g. https://pypi.org/project/google-genai/), and the Gemini engine flow requires this dependency to run.

Issues (2)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 5, 2026, 03:56 AM
Issues
2
Security Audit — snyk — VideoTranscribe