agent-wallet-policy

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Documentation and external references are sourced from official Circle domains and GitHub repositories.
  • [SAFE]: The skill incorporates strong security guardrails by explicitly instructing the agent never to handle, store, or request sensitive authentication data like One-Time Passwords (OTPs).
  • [SAFE]: Sensitive state-changing operations, such as setting or resetting spending limits, are implemented as 'human-in-the-loop' workflows. The agent is restricted to providing the user with a verbatim command to be executed manually in the user's local terminal, ensuring the agent remains outside the trust boundary for authentication.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 11:38 AM