aster-trading
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill communicates with legitimate Aster DEX API and WebSocket endpoints (fapi.asterdex.com, sapi.asterdex.com, and fstream.asterdex.com) to fetch market data and execute authenticated requests.
- [COMMAND_EXECUTION]: Provides instructions for users to set up the
aster-mcpCLI tool for account configuration. These commands are intended for local setup and are not executed autonomously by the agent. - [DATA_EXFILTRATION]: No unauthorized data exposure detected. The skill uses a secure configuration pattern where API keys are stored in an encrypted format within the user's home directory (~/.config/aster-mcp/).
- [PROMPT_INJECTION]: The skill explicitly defines mandatory safety rules that prevent the agent from placing orders without a preview and user confirmation, effectively enforcing human-in-the-loop oversight.
- [SAFE]: Regarding indirect prompt injection (Category 8), while the skill ingests untrusted market data (e.g., K-lines, ticker info) from external APIs, it implements robust boundary markers in its instructions that require the agent to pause for user approval before taking any significant action based on that data.
Audit Metadata