The Agent Skills Directory

[SAFE]: The skill operates entirely locally using a static JSON data file (data/strategy_patterns.json) and a Python script (scripts/strategist.py) to provide architectural advice.
[COMMAND_EXECUTION]: The skill uses argparse to handle user input for architectural topics and scalability tiers. These inputs are used for dictionary lookups and comparison logic within the script, with no evidence of shell injection or unsafe subprocess execution.
[DATA_EXPOSURE]: The script reads from a local JSON file included with the skill. There are no network operations, file writes, or access to sensitive system paths.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied strings for topic searches. However, the script implements strict validation via choices in argparse and simple string matching against a hardcoded data structure, preventing any malicious instruction override.

system-strategist