telnyx-webrtc-client-ios

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt's examples and instructions show credentials (SIP username/password, JWT token, APNS device token) being placed directly as string literals and tell the developer to "save token for use in TxConfig", which would require an agent to accept and embed secret values verbatim in generated code—an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The SKILL.md "AI Agent Integration" and PushKit sections show the client performing anonymousLogin and subscribeToTranscriptUpdates (and reading PushKit payload.metadata) which ingest remote, user-generated transcripts and push metadata from the Telnyx backend/remote callers—untrusted third-party content that the agent reads and could materially influence subsequent behavior.