The Agent Skills Directory

[SAFE]: The skill utilizes a set of internal tools prefixed with Amplitude: (e.g., search, get_event_properties, query_dataset). All data operations are performed within the vendor's platform environment.
[SAFE]: The instructions prioritize a 'Discovery First' approach, instructing the agent to broadly search and evaluate results before building definitions. It explicitly requires the agent to verify data volume and explain selection rationale to the user, providing a human-in-the-loop safety mechanism against potential data misinterpretation.
[SAFE]: There are no signs of malicious patterns, such as prompt injection, hardcoded credentials, data exfiltration to third-party domains, or unauthorized command execution. The JSON templates provided are standard for API interaction and do not involve obfuscated content or remote code execution.

create-chart