The Agent Skills Directory

[SAFE]: The skill performs static validation of local 1C configuration XML files. It is designed to detect structural errors and inconsistencies in metadata objects.
[COMMAND_EXECUTION]: The Python script scripts/meta-validate.py uses subprocess.call to implement batch processing by recursively calling itself. The command is constructed as a list, and it does not use shell=True, preventing shell injection vulnerabilities.
[DATA_EXFILTRATION]: No network operations, credential harvesting, or access to sensitive user files (such as SSH keys or cloud provider configurations) were detected. File access is restricted to the local 1C configuration files for validation purposes.
[PROMPT_INJECTION]: No malicious instructions, safety bypass attempts, or system prompt extraction patterns were found in the skill's metadata or instructional content.

meta-validate