The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the max benchmark CLI tool via pixi run to perform performance testing. This is the intended purpose of the skill and involves standard local command execution.\n- [DATA_EXFILTRATION]: Network operations are confined to checking the status of a local server at http://localhost:8000. There are no indications of data being sent to external or unauthorized endpoints.\n- [EXTERNAL_DOWNLOADS]: The tool supports using datasets like sharegpt from Hugging Face Hub for benchmarking. These are standard resources from a well-known service and do not represent a security risk.\n- [CREDENTIALS_UNSAFE]: No sensitive information, such as API keys or private tokens, is hardcoded or requested by the skill instructions.

benchmark-model