compose

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow explicitly requires verifying against "live source code" by fetching public repositories (see "Source Code Verification" and Tier 2: raw GitHub URLs like https://raw.githubusercontent.com, gh api repos/... and android.googlesource.com) so the agent will ingest and act on untrusted, user-facing third-party content from the open web.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs fetching live source at runtime (e.g., https://raw.githubusercontent.com/androidx/androidx/androidx-main/{path} and related raw GitHub / android.googlesource.com URLs) to verify behavior and inject that content into the agent's context, so these external URLs are used at runtime and can directly control the agent's prompts/responses.