saas-metrics-coach

This skill is internally coherent and appears to do what it claims: local calculation, benchmarking, and recommendations. Before installing or using it: (1) Review the Python scripts yourself or run them in an isolated environment if you have any doubt — they use only the Python standard library and contain no network calls. (2) Be cautious about pasting very sensitive or identifying business data into the agent; the skill needs numeric revenue/customer inputs but does not require secrets. (3) Note small implementation details (e.g., gross-margin handling via percent vs decimal defaults and some input validation relying on truthy checks) — verify outputs for edge cases. (4) If you will allow the agent to invoke the skill autonomously, ensure your agent policies / prompts do not cause it to include or exfiltrate unrelated context; the skill itself does not contact external endpoints. If you want higher assurance, run the scripts locally on a sample dataset first and confirm results match your expectations.