land-and-deploy

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly reads and interprets untrusted, user-generated content (e.g., PR body via "gh pr view --json body" in Step 3.5c and arbitrary production URLs via the $B goto canary steps) and uses that content to build readiness reports and drive merge/deploy/revert decisions, so third‑party content can materially influence actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's SETUP step includes running curl -fsSL https://bun.sh/install | bash which downloads and executes remote code at runtime as part of the required build/install path for the browse tool, so it directly executes external code and is a runtime dependency.