dripping-faucet
Pass
Audited by Gen Agent Trust Hub on May 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security vulnerabilities were detected. The skill implementation follows robust security best practices: \n
- Secure Credential Management: The skill mandates using environment variables or the Radius CLI wallet convention (encrypted keystores) to prevent private key exposure. It explicitly prohibits logging keys or passing them as insecure CLI arguments. \n
- Input Validation: Instructions require mandatory address validation (via
isAddressor regex) before initiating requests, which prevents processing of malformed data. \n - Indirect Prompt Injection Mitigation: The skill defines a strict trust boundary for external API responses. It instructs the agent to treat all response content as data only and to ignore any instructions embedded in informational fields such as 'message' or 'instructions'. \n
- Credential Safety: Security rules require sensitive files (like
.envor.radius/) to be listed in.gitignoreto prevent accidental exposure in version control.
Audit Metadata