msverl-daily-regression-triage

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow (sync_repos.py) clones public repositories (https://github.com/verl-project/verl.git and https://gitcode.com/Ascend/MindSpeed.git) and list_recent_commits.py reads commit messages and touched files which rank_candidate_commits.py and triage_msverl_regression.py then use to drive candidate ranking and fix-direction decisions, so untrusted, user-generated repo content is ingested and can materially influence actions.