The Agent Skills Directory

[SAFE]: The skill documentation and referenced API endpoints align with legitimate administrative tasks for Confluence. No obfuscation, prompt injection, or data exfiltration attempts were found.
[COMMAND_EXECUTION]: The skill utilizes a confluence CLI tool to perform its operations. These commands are focused on permission management (get, add, remove) and follow expected patterns for such a utility.
[DATA_EXPOSURE]: While the skill manages sensitive access controls, it does not demonstrate any patterns of exfiltrating this data to unauthorized external services. All network operations described are directed at official Confluence API endpoints.

confluence-permission