aomi-transact
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates a high degree of security awareness and maturity. It defines a comprehensive permission manifest based on the OWASP Agentic Skills Top 10 framework, including restricted shell access and network allowlisting.
- [COMMAND_EXECUTION]: All shell operations are strictly scoped to the official 'aomi' CLI and version-pinned 'npx' invocations. This adherence to least privilege prevents the agent from executing unauthorized system commands.
- [EXTERNAL_DOWNLOADS]: Remote dependencies are limited to the vendor's own official client package hosted on NPM, a well-known service. The use of version pinning (@0.1.30) ensures integrity and prevents unintended updates from unknown sources.
- [PROMPT_INJECTION]: The skill incorporates specific defensive instructions to handle untrusted input, particularly during transaction construction. It mandates 'drain-vector' checks during simulation to detect and surface attempts to redirect funds to non-user-owned addresses.
Audit Metadata