skill-rails-upgrade

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests public third‑party webpages to drive decisions — e.g., Step 5 uses WebFetch to read the Rails Upgrade Guide (https://guides.rubyonrails.org/upgrading_ruby_on_rails.html) and Step 6 reads railsdiff.org (https://railsdiff.org/{current_version}/{target_version}), and those fetched pages are parsed and summarized to determine upgrade actions and file merges.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches and uses external pages at runtime—https://guides.rubyonrails.org/upgrading_ruby_on_rails.html and https://railsdiff.org/{current}/{target}—and instructs the agent to extract and summarize those pages to drive its upgrade guidance, so the fetched content directly controls the agent's instructions and is required for the skill to operate.