The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute lifecycle management commands via the UiPath CLI (uip) and the uv package manager. These operations include project initialization, dependency synchronization, and local execution of agent logic, all of which are standard for this development workflow.
[EXTERNAL_DOWNLOADS]: Fetches the official UiPath CLI tool (@uipath/cli) from NPM and relevant SDK packages (such as uipath-langchain and uipath-llamaindex) from PyPI. These downloads originate from the vendor's own verified infrastructure.
[CREDENTIALS_SAFE]: Employs secure authentication practices by checking for existing tokens in .env files and requiring interactive user input for sensitive environment, organization, and tenant details before running authentication commands.
[DYNAMIC_EXECUTION]: As its core functionality, the skill generates and executes Python scripts based on user requirements. This behavior is confined to the local project environment for the purpose of building and testing AI agents.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of agents that ingest external data. While this presents an inherent surface for indirect prompt injection, the skill incorporates the UiPath evaluation framework (including smoke-test.json) to validate agent behavior against expected outcomes.

uipath-coded-agents