meta-repo

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow fans out meta commands across child repos listed in a tenant-controlled .meta file, and meta exec/meta git necessarily operate on those repos—meaning outsider-authored text from untrusted remotes (e.g., cloned repo contents/commits or command outputs) is ingested/executed as part of the workflow.