youtube-apify-transcript

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted transcript data from the web.
  • Ingestion points: Transcripts are retrieved from the YouTube platform via external API calls in scripts/fetch_transcript.py. The script also reads user-controlled files when using the --batch flag.
  • Boundary markers: Absent. The tool outputs transcript data as raw text or structured JSON without specific delimiters to isolate it from the agent's context or instruction set.
  • Capability inventory: The script possesses network access via the requests library and file system capabilities for reading batch inputs and writing to a local cache directory.
  • Sanitization: Absent. No filtering or validation is performed on the retrieved caption content before it is passed to the output stream.
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to api.apify.com to trigger and retrieve results from a web scraping service. Apify is a well-known and legitimate platform for data extraction, and this interaction is necessary for the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 06:03 PM