injective-trading-tokens
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its ingestion of external blockchain registry data and user-supplied parameters for financial actions.\n
- Ingestion points: The skill retrieves token data from the on-chain registry via the
token_metadatatool and accepts user-provided denoms and wallet addresses as documented inSKILL.md.\n - Boundary markers: No specific delimiters or safety instructions are defined in the skill documentation to separate potentially untrusted blockchain data from agent instructions.\n
- Capability inventory: The skill enables high-impact capabilities such as
transfer_send,subaccount_deposit, andsubaccount_withdrawfor moving funds (defined inSKILL.md).\n - Sanitization: The instructions note that the server handles decimal conversion internally but do not detail validation or sanitization for external strings like addresses or denoms.
Audit Metadata