The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the cloud CLI tool to perform infrastructure and application management tasks. Instructions emphasize using the -n flag for non-interactive execution and runtime discovery of flags.
[REMOTE_CODE_EXECUTION]: Includes documentation for cloud tinker and cloud command:run, which allow the agent to execute PHP code and shell commands on remote cloud environments as a core administrative feature.
[EXTERNAL_DOWNLOADS]: Fetches official documentation from https://cloud.laravel.com/docs/llms.txt using the WebFetch tool to improve the accuracy of its responses.
[DATA_EXFILTRATION]: Identifies that the CLI stores sensitive authentication tokens and preferences in ~/.config/cloud/config.json for persistent access to the Laravel Cloud API.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted documentation from an external URL (https://cloud.laravel.com/docs/llms.txt). The skill possesses the capability to execute shell commands via the cloud CLI. No explicit boundary markers or sanitization logic for this external content are documented.

deploying-laravel-cloud