voice-ai-development
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes patterns for ingesting and processing real-time audio transcripts and webhook events, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: The
vapi_webhookfunction inSKILL.mdprocesses JSON payloads from external Vapi calls, and theon_transcriptcallback inSKILL.mdhandles text results derived from live audio streams via Deepgram. - Boundary markers: The code examples provided do not demonstrate the use of delimiters or specific system instructions designed to prevent the model from following malicious commands hidden within the audio transcripts or webhook data.
- Capability inventory: The skill demonstrates the ability to execute network operations (via API clients) and perform tool-like actions (e.g., the
check_orderfunction call) based on the processed inputs. - Sanitization: There is no evidence of data sanitization or validation for the external content before it is processed by the AI agent logic.
Audit Metadata