The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the platform's CLI utility to perform market operations.
Evidence: In scripts/dog_config.py, the mcporter_call function uses subprocess.run to execute commands. These commands are constructed using a list structure and JSON-serialized arguments, which prevents shell injection.
Evidence: README.md and SKILL.md provide a bash loop example (nohup bash -c 'while true; do python3 ...') for users to run the scanner in a detached background process.
[EXTERNAL_DOWNLOADS]: The skill's setup process involves downloading its core logic and configuration from a remote source.
Evidence: README.md and SKILL.md include curl commands to fetch files from the author's official repository at github.com/Senpi-ai/senpi-skills. These resources are necessary for the skill's deployment and originate from the vendor's infrastructure.
[SAFE]: The skill follows security best practices for state management and configuration handling.
Evidence: Secret management is handled via environment variables and configuration placeholders (${WALLET_ADDRESS}) rather than hardcoded credentials.
Evidence: Local state tracking for trade limits is managed via atomic file writes in scripts/dog_config.py to ensure data integrity.

dog-strategy