write-atomic-tasks
Pass
Audited by Gen Agent Trust Hub on Feb 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and a utility script to help agents write more precise tasks. No malicious code or instructions were found.
- [COMMAND_EXECUTION]: The provided Python script
scripts/validate_task.pyis a benign tool that uses standard libraries (argparse,re) to validate text input. It does not perform any dangerous operations like system command execution or unauthorized file modifications. - [DATA_EXFILTRATION]: No hardcoded credentials or sensitive data access patterns were found. The script reads files only when explicitly directed by the user via command-line arguments.
- [PROMPT_INJECTION]: The instructions in
SKILL.mdare focused on formatting and task decomposition rules. They do not attempt to override system prompts or bypass safety filters. - [REMOTE_CODE_EXECUTION]: No patterns of remote script downloading or dynamic code execution were detected across any of the files.
Audit Metadata