Transcription Automation

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Potential surface for indirect prompt injection through untrusted media inputs.
  • Ingestion points: Transcribes untrusted audio and video content from Zoom, YouTube, and RSS feeds as defined in the integration workflows (SKILL.md).
  • Boundary markers: Lacks delimiters or instructions to the model to ignore potential commands spoken within the transcribed media content.
  • Capability inventory: The skill has the capability to distribute processed transcripts and summaries to external services including Notion, Slack, and email (SKILL.md).
  • Sanitization: The skill does not perform sanitization or filtering of transcribed text for adversarial instructions before it is processed by the summarization model.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 05:19 AM