release-notes

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads local git commit messages and lists merged GitHub pull requests, which can contain outsider-authored free text from untrusted contributors.

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

• Potentially malicious external URL detected (high risk: 0.70). The installation command in SKILL.md references an external GitHub repository URL to add the skill, but the target repository is personal and arbitrary rather than an established official publisher, presenting a weak third-party runtime dependency.