software-engineer

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to "Read https://wazoo.dev's JSON-LD graph" (an external public website) to synchronize company details, which requires ingesting untrusted third-party content that can materially influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs on-load fetching of https://wazoo.dev's JSON-LD graph to "synchronize with the company", which is a runtime fetch of external content that can directly alter the agent's context/prompts.