The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the uip CLI tool and curl commands to manage Orchestrator folders, assets, and processes. It also includes instructions for managing UiPath Studio instances using rpa-tool.
[DATA_EXFILTRATION]: Instructions direct the agent to read the ~/.uipath/.auth file to extract access tokens. These tokens are used to authenticate network requests made with curl to official UiPath Orchestrator endpoints.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to install the @uipath/cli package from the npm registry and various official CLI tool extensions such as @uipath/resources-tool and @uipath/test-manager-tool.
[PROMPT_INJECTION]: The skill processes data from external UiPath services (Orchestrator and Integration Service) which serves as an attack surface for indirect prompt injection.
Ingestion points: Output from uip is resources list, uip or folders list, and uip tm project list referenced in agent-workflow.md, resources.md, and orchestrator-guide.md.
Boundary markers: None identified; there are no instructions to the agent to distinguish between tool output and system instructions.
Capability inventory: The skill has permission to execute shell commands (Bash) and write to the file system (Write) as specified in the allowed-tools metadata.
Sanitization: No evidence of data sanitization or validation of the retrieved external content before processing.

uipath-platform