The Agent Skills Directory

[SAFE]: The skill consists of documentation and reference patterns for the Modal Python SDK, aimed at managing isolated execution environments.
[COMMAND_EXECUTION]: While the skill facilitates command execution, it is explicitly designed to do so within isolated Modal Sandboxes. The instructions emphasize using modal.Sandbox.create and sandbox.exec for this purpose.
[DATA_EXFILTRATION]: No patterns of data exfiltration were found. The skill includes guidance on using modal.Secret for handling credentials securely rather than hardcoding them.
[EXTERNAL_DOWNLOADS]: The skill does not perform any unauthorized external downloads. It relies on the well-known modal Python library for its operations.
[PROMPT_INJECTION]: No attempts to override agent behavior or safety guidelines were detected in the instructions or metadata.
[SAFE]: The provided scripts/smoke_test.py is a legitimate diagnostic script used to verify that the local environment can correctly communicate with the Modal API and perform basic sandbox operations.

modal-sandbox