Contract Addresses

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/addresses/SKILL.md at install/runtime, which fetches outsider-authored free text from the public web (EthSkills) and can be ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The catalog explicitly instructs fetching and reading https://ethskills.com/addresses/SKILL.md at runtime (curl in install/setup and "Read ... before using addresses"), and the remote document provides the contract addresses the agent must use—i.e., external content directly controls agent behavior—so this is a runtime dependency that can alter prompts/outputs.