uipath-human-in-the-loop

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill reads the file ~/.uipath/.auth to retrieve authentication credentials. This is used to authenticate requests to official UiPath cloud services (e.g., alpha.uipath.com) for resource discovery and configuration. As the skill is authored by the vendor (UiPath) and operates within its own platform architecture, this is considered legitimate operational behavior.
  • [COMMAND_EXECUTION]: Utilizes the uip CLI tool and standard system commands (find, sed, rsync) via the Bash tool to perform project scaffolding, file discovery, and flow validation. It also uses node -e for local UUID generation.
  • [PROMPT_INJECTION]: Acts as a surface for indirect prompt injection by processing external data such as .flow files and user-defined business schemas. The skill incorporates multiple manual confirmation steps, requiring the user to explicitly approve schemas and changes before they are committed to project files.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 06:08 AM
Security Audit — agent-trust-hub — uipath-human-in-the-loop