voxflow

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md's "Article → Audio Briefing" workflow explicitly instructs the agent to "Fetch/read the content" from a URL (public webpages) and then summarize and synthesize it, meaning the agent ingests untrusted third‑party web content that can influence its outputs and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent at runtime to add an MCP using the URL "https://api.voxflow.studio/api/mcp" (via claude mcp add voxflow https://api.voxflow.studio/api/mcp), which registers remote tool endpoints that alter agent behavior and enable remote-executed actions, so it is a runtime external dependency that can directly control the agent.