deepgram-recipes

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches code snippets and documentation from Deepgram's official GitHub repository (github.com/deepgram/recipes). This is a well-known technology service and the source is considered trustworthy.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes content from an external repository, creating a potential surface where embedded instructions could influence the agent. However, the risk is minimal given the trusted nature of the source.
  • Ingestion points: Retrieves documentation and code from github.com/deepgram/recipes.
  • Boundary markers: The instructions do not define explicit delimiters to isolate external content from the agent's system prompt.
  • Capability inventory: The skill is limited to documentation retrieval and code presentation; it does not involve automated command execution, file writing, or network operations on the ingested data.
  • Sanitization: No specific filtering or sanitization of the retrieved content is defined in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:55 AM
Security Audit — agent-trust-hub — deepgram-recipes