agent-smith

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

• Third-party content exposure detected (medium risk: 0.30). The workflow explicitly reads GitHub issues using gh issue list and prompts workers to read issue bodies/comments via gh, exposing the agent to outsider-authored free text.

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

• Potentially malicious external URL detected (high risk: 0.70). The README instructs running npx skills add aslamdoctor/skills --skill agent-smith, which downloads and executes arbitrary third-party code from an unverified public GitHub repository at runtime without cryptographic verification or pinning.