The Agent Skills Directory

[SAFE]: The skill facilitates legitimate interaction with the inference.sh platform using its dedicated command-line interface. All external links point to official documentation and repositories associated with the service.
[SAFE]: The YAML frontmatter implements the principle of least privilege by using the allowed-tools field to restrict the agent's shell access specifically to the belt command, preventing arbitrary command execution.
[SAFE]: No malicious patterns such as obfuscation, credential harvesting, or unauthorized data exfiltration were detected in the instructions or examples provided.

p-image