awx-best-practices
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the
airwallexCLI binary for various operations, including listing, getting, updating, and deleting resources such as invoices, cards, and beneficiaries. - [COMMAND_EXECUTION]: Automated authentication management is described, where the agent is told to execute
airwallex auth loginif an authorization error occurs, after confirming the target environment with the user. - [SAFE]: The skill incorporates strong safety guardrails, requiring the use of
--dry-runand explicit user approval (via the--confirmflag) for all potentially destructive or high-impact actions. - [SAFE]: Instructions explicitly forbid the fabrication of data and require the agent to stop and ask the user if any required information is missing or ambiguous.
Audit Metadata