Standards

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/standards/SKILL.md and then instructs the agent to read that fetched page, so the LLM ingests outsider-authored free text from EthSkills’ public web content at runtime.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's install/setup explicitly runs a runtime fetch of remote content that can control agent behavior—e.g., "curl -s https://ethskills.com/standards/SKILL.md" (and a suggested "claude plugin install https://github.com/austintgriffith/ethskills") which would supply external instructions or code used at runtime.