gate-exchange-simpleearn
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious prompt injection or behavior override patterns were detected. The skill uses strong internal instructions to prevent the agent from attempting to bypass service restrictions, specifically regarding the disabling of financial transactions.
- [DATA_EXFILTRATION]: The skill exhibits safe data handling practices. It contains explicit rules against logging or exposing API keys and sensitive account balances. It correctly identifies authentication failures and directs users to configure their own keys within the MCP environment rather than requesting them in-chat.
- [EXTERNAL_DOWNLOADS]: The skill references the 'gate-mcp' toolset. These references point to the author's own official GitHub repository (github.com/gate/gate-mcp), which is considered a trusted vendor resource for this skill.
- [COMMAND_EXECUTION]: No dangerous system command execution or shell scripting was found. The skill operates via structured MCP tool calls to the Gate API.
- [REMOTE_CODE_EXECUTION]: No patterns of remote code execution, dynamic code loading, or untrusted script evaluation were identified.
Audit Metadata