gbrain-upgrade

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime flow ingests OUTSIDER-authored free text from gbrain self-upgrade --check-only --json, specifically the remote changelog_diff and release_url fields (fetched from upstream release notes), which are then summarized into plain bullets and included in the LLM context for prompting the operator.