deepgram-webhooks-events
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements endpoints to receive and process transcription payloads from an external webhook. This creates a surface for indirect prompt injection, as malicious text transcribed from audio could influence the behavior of an AI agent if the resulting transcript is used in subsequent prompts without proper sanitization.
- [EXTERNAL_DOWNLOADS]: The skill references several standard Node.js packages (@deepgram/sdk, express, ioredis, uuid) required for the webhook server and client implementation, which are fetched from the official NPM registry.
Audit Metadata