demo-skill

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and summarizes content from the public GitHub profile at https://github.com/slezakattack, which is user-generated/open third-party content that the agent will read and interpret.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to fetch and summarize content from https://github.com/slezakattack, so that external content directly controls the agent's prompts/output.