vllm-plugin-fl-setup-flagos

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md installation workflow (Steps 2–4) explicitly instructs fetching and installing code from public GitHub repositories (e.g., git clone https://github.com/flagos-ai/vllm-plugin-FL, FlagGems, FlagCX) so the agent will ingest and execute untrusted, user-controlled third-party content that can alter installation/configuration actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs runtime git clones and pip installs that fetch and execute remote code from external repositories (e.g. https://github.com/flagos-ai/vllm-plugin-FL, https://github.com/flagos-ai/FlagGems, https://github.com/flagos-ai/FlagCX.git) and a custom pip index (https://resource.flagos.net), which are required dependencies and thus execute remote code at runtime.