ce-promote

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

• Third-party content exposure detected (high risk: 0.75). Runtime LLM context can include outsider-authored text from gh pr view --json title,body,url (PR title/body written by non-operating-user contributors) and/or git diff/changelog/commit messages derived from those changes, which the skill then summarizes and uses as the basis for drafting.