cortex-router

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

• Third-party content exposure detected (high risk: 0.80). The skill routes prompts to the Cortex Code CLI and its required workflow (Step 3 in SKILL.md) explicitly supports a "RESEARCH" security envelope and lists a web_fetch tool in references/cortex-cli-reference.md, indicating the agent may fetch and read arbitrary public web content (untrusted third-party pages) which could influence subsequent tool use.