bmad-walkthrough

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

• Third-party content exposure detected (medium risk: 0.30). The skill reviews a target such as a PR or commit, which can include outsider-authored pull request text and commit messages.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill references an external repository (github:bmad-code-org/BMAD-METHOD/skills) and package manager (npx skills add bmad-code-org/BMAD-METHOD --skill bmad) to install dependencies at runtime, representing an unofficial remote dependency.