video-transcriber
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text from video transcripts which could contain embedded instructions aimed at overriding agent behavior.
- Ingestion points: Transcript segments are retrieved in
video_transcriber.pyfrom external platform APIs. - Boundary markers: While the tool provides Markdown and JSON structure, it does not include explicit delimiters or instructions for the agent to ignore potentially malicious commands within the transcript content.
- Capability inventory: The skill performs network requests and has the ability to write to the local filesystem for output storage.
- Sanitization: No sanitization or escaping of the retrieved transcription text is performed.
- [EXTERNAL_DOWNLOADS]: Fetches video metadata and subtitle tracks from YouTube and noembed.com. These are part of the core functionality. The skill correctly validates hostnames to ensure requests are limited to supported video platforms.
Audit Metadata