broadcast

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

• Third-party content exposure detected (high risk: 0.90). The skill auto-discovers local repos and explicitly runs git fetch / git pull against each repo's remote and then "analyze[s] the repo structure" and applies changes/runs tests/builds, which means it ingests and acts on untrusted, user-generated repository content from external remotes (e.g., public GitHub/git remotes) that can materially influence agent actions.