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

shark

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The core logic in scripts/shark_config.py uses subprocess.run to interact with mcporter, a CLI tool used for calling platform-specific tools. This is the standard method for API interaction on the Senpi platform and follows best practices by using argument lists rather than shell strings.
  • [COMMAND_EXECUTION]: The bootstrap instructions in SKILL.md use sed to safely update local configuration files (runtime.yaml) with user-provided wallet addresses and Telegram identifiers.
  • [SAFE]: The skill implements automated trading logic including scoring, risk management, and cooldowns. It relies on an external platform dependency (senpi-trading-runtime) for position tracking and exit management, which is consistent with its stated purpose.
  • [SAFE]: Data handling is restricted to market telemetry and strategy state stored within the workspace. No evidence of unauthorized data exfiltration or credential harvesting was found.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 05:54 PM