telnyx-ai-assistants-ruby

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

• Third-party content exposure detected (high risk: 0.80). The skill's docs (SKILL.md and references/api-details.md) expose a dynamic_variables_webhook_url and other configurable webhooks/import endpoints (e.g., dynamic_variables_webhook_url, MCP server url, import from external provider) which cause the assistant to fetch and consume data from arbitrary third-party URLs that can alter assistant behavior, so untrusted external content could indirectly inject instructions.