The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a Python script (scripts/run_strategy.py) to process strategy requests. The script is executed via the bash shell as defined in the manifest and interface configuration.- [DYNAMIC_EXECUTION]: The execution script dynamically modifies the Python search path (sys.path) to include a local directory for importing the geo_seo_hub module. This is a common pattern for local tool development to reference internal libraries within the same project structure.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a file-backed JSON request. It mitigates risks by requiring explicit fidelity checks (risk, brand, constraint) and maintaining an offline boundary where all external evidence must be manually verified.- [SAFE]: The skill configuration explicitly forbids network access and autonomous execution of external actions, effectively minimizing the potential attack surface.

geo-strategy