write-script-rlang

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides legitimate operational instructions for using the wmill CLI to manage R scripts. It explicitly instructs the agent to favor local previews over direct workspace deployments to prevent overwriting stable versions with untested changes, promoting a secure development workflow.
  • [COMMAND_EXECUTION]: The skill utilizes the wmill CLI tool for script execution and management. These commands are standard for the developer platform associated with the skill author (windmill-labs) and are used to provide the skill's primary functionality.
  • [PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection by mapping user instructions to command execution. However, the instructions provide specific logic to sanitize this interaction by deriving arguments from script metadata rather than raw input. Ingestion points: User requests via the agent. Boundary markers: Absent. Capability inventory: wmill CLI (local execution, deployment, file sync). Sanitization: Arguments are selected based on script-defined parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 06:29 AM
Security Audit — agent-trust-hub — write-script-rlang