The Agent Skills Directory

[SAFE]: The skill operates as an advisory tool for clinical evaluation design. Technical review of scripts/check_mllm_eval_completeness.py shows it is a task-aware linter using standard library regex for presence checking, with no external dependencies or remote execution paths.\n- [SAFE]: No network exfiltration, remote code execution, or credential management issues were detected. Auditing logic is performed locally on user-provided protocol drafts without dangerous interpolation of untrusted content.

mllm-eval