smart-accounts
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill uses the fdx CLI tool to perform blockchain operations. Command execution is restricted through explicit tool definitions for specific subcommands such as status, deploySmartAccount, and manageSmartAccountOwnership.
- [Indirect Prompt Injection] (SAFE): The skill ingests untrusted data into command flags.
- Ingestion points: User-provided addresses and chain identifiers passed to CLI arguments in SKILL.md.
- Boundary markers: Delimiters are absent; inputs are interpolated directly into command strings.
- Capability inventory: The skill can deploy contracts and modify ownership structures on EVM chains.
- Sanitization: While no sanitization is defined in the markdown, the skill requires human confirmation for all sensitive ownership changes to mitigate the risk of malicious input injection.
Audit Metadata