The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses __import__ in scripts/portfolio_commands/__init__.py to dynamically load internal modules for various subcommands. This is a modular programming pattern where the module paths are hardcoded in a registry and are not influenced by external or untrusted input.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external market data from a Yahoo Finance wrapper and processes user-supplied investment memos. The attack surface is effectively managed by using structured data formats and standard command-line argument parsing via argparse, which validates inputs before they are processed by the core logic.
[COMMAND_EXECUTION]: The skill entry point run_portfolio.py is executed via the Bash tool to process user instructions. It routes commands to specific Python functions based on arguments, which is the intended and documented functionality of the agent skill.
[DATA_EXPOSURE]: The skill reads and writes to a local portfolio.csv file to track holdings. This data access is restricted to the skill's specific data directory and is necessary for its stated purpose of portfolio management.

stock-portfolio