injective-trading-autosign
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a session-based trading authorization model using the Injective blockchain's AuthZ primitive. It defines clear scopes for permissions, focusing on trading-related message types while blacklisting sensitive operations.
- [SAFE]: The instructions include safety checks for AI behavior, such as a requirement to reject requests for granting 'MsgSend' or 'MsgWithdraw' permissions, which helps prevent unauthorized asset transfers.
- [SAFE]: All external references, including RPC endpoints (e.g., sentry.evm-rpc.injective.network) and SDK dependencies (@injectivelabs/sdk-ts), are official resources provided by the vendor, Injective Labs.
- [SAFE]: The activity definitions for 'authz_grant' and 'authz_revoke' use appropriate placeholders for sensitive inputs like passwords, following standard practices for keystore-based transaction signing in a local environment.
Audit Metadata