assemblyai-transcription
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a documentation resource for AssemblyAI's transcription services, recommending secure practices like environment variables for API keys.
- [SAFE]: It provides detailed guidance on data privacy, including TTL settings and model-training opt-outs.
- [SAFE]: All links reference official AssemblyAI documentation on their primary domain.
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing untrusted media, creating an attack surface for indirect prompt injection. This is mitigated by guidance on human review and redaction.
- Ingestion points: External media files.
- Boundary markers: Manual verification requirements.
- Capability inventory: API calls, file writes, and LLM processing.
- Sanitization: PII redaction and content moderation guidance.
Audit Metadata