The Agent Skills Directory

[SAFE]: The skill implements its trading strategy using official platform tools (mcporter, openclaw) for market analysis and operational tasks.
[COMMAND_EXECUTION]: The script scripts/hydra_config.py utilizes subprocess.run to execute local platform utilities. The implementation is secure as it uses a command list format instead of a shell string and encodes arguments into JSON to prevent injection.
[DATA_EXFILTRATION]: No unauthorized network operations or data transmission to external domains were identified. Data access is restricted to market and account information via the platform's internal MCP mechanisms.
[PROMPT_INJECTION]: Instructions in SKILL.md define operational rules and configuration steps for the agent without attempting to override safety filters or extract system prompts.

hydra-strategy