response-compliance

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "How It Works" section explicitly says the test script clones the public GitHub repo openresponses/openresponses into scripts/openresponses-compliance/ and runs its CLI test runner, meaning the agent fetches and executes untrusted third‑party repository content that can alter runtime behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's compliance-test script explicitly clones and runs the remote repository https://github.com/openresponses/openresponses at runtime, which fetches and executes remote code (the repo's CLI test runner), so the external URL directly controls execution.