The Agent Skills Directory

[SAFE]: The skill's behavior is consistent with its stated purpose of parsing local XML files for role auditing. No malicious patterns, external downloads, or data exfiltration attempts were detected.
[COMMAND_EXECUTION]: The skill utilizes PowerShell to execute a local script (role-info.ps1). This is a standard and safe use of the command-line interface for structured data processing within the agent's environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from user-provided XML files, which presents a surface for indirect prompt injection if those files contain malicious instructions.
Ingestion points: scripts/role-info.ps1 reads data from Rights.xml and associated metadata XML files.
Boundary markers: The skill output does not currently utilize specific delimiters or warnings to isolate the parsed content from the agent's instruction context.
Capability inventory: The skill has the capability to read files from the filesystem, write results to files, and execute PowerShell scripts.
Sanitization: Content is extracted from XML nodes using standard .NET XML libraries without additional sanitization before being presented to the agent.

role-info