The Agent Skills Directory

[SAFE]: The skill uses the subprocess.run function to interact with the mcporter command-line utility for market data and trading. This is standard platform behavior for skills operating within the Senpi ecosystem.
[SAFE]: Sensitive information such as wallet addresses and strategy IDs are sourced from environment variables or configuration files with proper placeholders, avoiding hardcoded secrets and adhering to security best practices.
[SAFE]: The skill ingests market data through mcporter in scripts/croc-scanner.py. While the data comes from external instruments, it is handled as structured JSON and numeric values are explicitly cast to floats. This process provides sufficient sanitization against indirect prompt injection vectors in the context of this strategy's capabilities.

croc-strategy