The Agent Skills Directory

[SAFE]: No malicious patterns detected. The skill instructions focus on structured data gathering and estimation modeling. No prompt injection overrides or safety bypass attempts were found in the skill definitions.
[COMMAND_EXECUTION]: The skill uses a local Python script scripts/factor_model.py to perform calculations. This execution is confined to the skill's own logic and uses standard libraries to process JSON-formatted factor data. The script includes comprehensive validation logic for numeric values, scenario names, and metadata, ensuring that processed data adheres to the expected schema.
[DATA_EXFILTRATION]: No network activity was observed. The skill and its associated scripts operate entirely on local data and user-provided inputs without transmitting information to external servers. The file reading capability in the script is restricted to loading the JSON payload for analysis.

fermi-estimation