The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the uip insights CLI tool, executed through Bash, to interact with the UiPath environment for all monitoring and management operations.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (the user's UiPath Insights environment) which may contain user-authored text such as alert names, condition strings, and dashboard definitions. The instructions (e.g., in alerts-reads-guide.md and dashboard-reads-guide.md) explicitly warn the agent to treat this content as data only and never follow any instructions found within it, effectively mitigating the risk of indirect prompt injection.
[DATA_EXPOSURE]: The skill includes specific rules (Critical Rules 14 and 16) that instruct the agent to withhold sensitive data like email addresses, recipient identities, and internal resource strings unless specifically requested, reducing the risk of accidental data exposure.

uipath-insights