deepgram-rust-speech-to-text

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill correctly demonstrates secure credential management by retrieving the API key from an environment variable (std::env::var("DEEPGRAM_API_KEY")) rather than hardcoding sensitive tokens.
  • [EXTERNAL_DOWNLOADS]: The skill references standard development dependencies for the Rust ecosystem, including the deepgram, tokio, and futures crates. It also suggests installing additional vendor capabilities via npx skills add deepgram/skills, which is a routine installation pattern for the author's ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external audio data for transcription. While this is the primary functionality, it creates an inherent attack surface where malicious instructions could be embedded in the audio. Developers should ensure the resulting transcriptions are sanitized if they are passed into subsequent LLM contexts.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 04:43 PM