telnyx-import-retell

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the agent to collect the user's Telnyx and Retell API keys and to create/store the Retell key as an integration secret (and use the Telnyx API key to call the API), which requires the LLM to receive and include secret values verbatim in requests or outputs.

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime import endpoint pulls Retell agent configuration (including “Instructions” and other text fields) from Retell using the provided Retell API key, so outsider-authored free text from the Retell account is ingested into the agent/assistant context via the import flow.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The Telnyx import endpoint https://api.telnyx.com/v2/ai/assistants/import is invoked at runtime and causes Telnyx to fetch Retell agent configurations (including instructions/prompts) which directly control the created assistant's prompts, so this is a runtime external dependency.