The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the uip CLI tool to interact with UiPath Studio and manage automation projects. This includes listing Studio instances, opening projects, and managing Integration Service connections.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation and update of NuGet packages through the uip rpa install-or-update-packages command. These dependencies are necessary for extending the capabilities of RPA workflows.
[REMOTE_CODE_EXECUTION]: The skill is designed to run and debug RPA workflows on the local machine using the uip rpa run-file command, which is a core feature for testing automations during the development process.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it reads and processes untrusted project data, such as .xaml files and configuration metadata, which could contain malicious instructions designed to manipulate the agent's logic.
Ingestion points: The skill reads project files including project.json, .xaml workflows, and local documentation folders.
Boundary markers: While the skill follows a structured multi-phase workflow, it does not specify explicit delimiters or "ignore instructions" warnings when processing the content of existing project files.
Capability inventory: The agent has the ability to read/write/edit files and execute code locally via the vendor's CLI.
Sanitization: There is no evidence of content sanitization for the data extracted from existing workflow files before it is processed by the AI.

uipath-rpa-workflows