validation-task-artifacts
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed of Markdown files providing guidelines, checklists, and templates for artifact review. It contains no executable scripts or binaries.\n- [PROMPT_INJECTION]: No instructions were found that attempt to bypass AI safety filters or override system-level directives.\n- [DATA_EXFILTRATION]: There are no commands or instructions that access sensitive files or transmit data to external servers.\n- [REMOTE_CODE_EXECUTION]: The skill does not download or execute scripts from remote sources.\n- [COMMAND_EXECUTION]: The skill documentation includes examples of verification commands (e.g.,
mkdir,dart run) as reference material for the artifacts it reviews, but it does not execute these commands itself.\n- [EXTERNAL_DOWNLOADS]: No external packages or remote dependencies are referenced for installation.
Audit Metadata