telnyx-voice-advanced-ruby

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents methods that ingest external data, such as join_ai_assistant and start_conversation_relay. These represent points where untrusted data could enter the agent's context.
  • Ingestion points: Parameters like participant, client_state, greeting, and custom_parameters in SKILL.md.
  • Boundary markers: None explicitly defined in the SDK examples.
  • Capability inventory: API interactions with Telnyx voice services.
  • Sanitization: Not addressed in the provided documentation snippets.
  • [OBFUSCATION]: The "Update client state" example uses a Base64-encoded string aGF2ZSBhIG5pY2UgZGF5ID1d. When decoded, this translates to "have a nice day =]", serving as a benign placeholder.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 04:55 PM