spot
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill handles sensitive user credentials (HTX API Key and Secret Key) which are transmitted to the official HTX API servers (api.huobi.pro) for authentication purposes. This behavior is the primary intended function of the skill and targets well-known official domains.
- [COMMAND_EXECUTION]: The documentation provides example shell scripts that utilize standard system tools like curl and openssl to perform cryptographic signing and network communication. These are provided as reference implementations for API interaction.
- [PROMPT_INJECTION]: The skill includes explicit safety instructions for the agent to follow, such as masking sensitive keys in outputs and requiring a user to provide a 'CONFIRM' message before executing live mainnet transactions.
- [SAFE]: Indirect prompt injection surface analysis. 1. Ingestion points: User provides API credentials via file upload as described in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: The skill uses curl to make authenticated network requests to financial endpoints. 4. Sanitization: The skill includes behavioral instructions to mask secrets and require transaction confirmation, which mitigates the risk of unauthorized actions triggered by malicious input.
Audit Metadata