Testing

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

• Third-party content exposure detected (high risk: 0.75). The required workflow uses curl -s https://ethskills.com/testing/SKILL.md at install/runtime, which fetches public web content authored by EthSkills (an outsider) and can be ingested as readable text into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's setup and install commands explicitly fetch and require the remote document https://ethskills.com/testing/SKILL.md at runtime (curl and "Read ... before writing tests"), and that fetched content is intended to be read and used to control the agent's prompts/instructions.