Orchestration

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

• Third-party content exposure detected (high risk: 0.85). The required workflow runs curl -s https://ethskills.com/orchestration/SKILL.md at install time, so the agent ingests readable free text fetched from an outsider-controlled public web source (EthSkills site) into its LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The catalog's install/setup explicitly runs "curl -s https://ethskills.com/orchestration/SKILL.md" and tells agents to "Read https://ethskills.com/orchestration/SKILL.md before scaffolding", so remote content is fetched at runtime and directly controls agent instructions.