triage

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

• Third-party content exposure detected (high risk: 0.90). The skill's reproduce.md and other workflow docs explicitly require fetching and ingesting public, user-generated content (e.g., "gh issue view" for GitHub issues, cloning GitHub repos, fetching gists via the GitHub API, and using stackblitz-clone for StackBlitz projects), which the agent must read and act on to reproduce/diagnose/fix issues, exposing it to untrusted third-party content that could carry indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The reproduce workflow explicitly runs commands to clone GitHub repos at runtime (e.g. "git clone https://github.com//.git "), which fetches remote code that will be installed/run as part of triage, so this is a runtime external dependency that can execute remote code.