skills/senpi-ai/senpi-skills/mamba/Gen Agent Trust Hub

mamba

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/mamba_config.py uses subprocess.run to call the platform's mcporter CLI utility. The implementation correctly uses list-based arguments to prevent shell injection, and the command is limited to authorized platform functions.
  • [DATA_STORAGE]: The skill manages internal state (such as trade counters and asset cooldowns) by writing to the authorized workspace path at /data/workspace/skills/mamba-strategy/state/. This is standard for persistence in this environment.
  • [PROMPT_INJECTION]: The SKILL.md file contains instructions regarding trading logic, leverage caps, and asset filtering. No patterns attempting to bypass agent safety filters or extract system prompts were observed.
  • [EXTERNAL_ACCESS]: Network interactions are handled via calls to the mcporter tool, which acts as a gateway for market data and exchange operations. No direct network requests or external code downloads from unknown sources are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 05:44 AM