tokens
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially contain malicious instructions, which is a standard risk for tools interacting with blockchain data and user-provided memos.
- Ingestion points: The
--token,--recipient, and--memoflags in thetransfercommand, as well as token metadata (names, symbols) retrieved from the Hiro API, are entry points for untrusted data. - Boundary markers: The skill instructions do not explicitly define boundary markers to prevent the agent from following instructions that might be embedded in transaction memos or token names.
- Capability inventory: The skill includes the capability to transfer assets (
tokensService.transfer) and list user holdings. These are the primary functions of the skill and are restricted by the requirement of an unlocked wallet. - Sanitization: Basic validation of address formats (starting with 'SP' or 'ST') and numeric conversion of amounts is recommended in the AGENT.md safety checks.
Audit Metadata