The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of CLI tools, including the cloud CLI, composer, npm, and php artisan, to manage application deployments and infrastructure.
Evidence: Core operations rely on executing shell commands such as cloud ship, cloud deploy, composer require, and php artisan migrate.
[DYNAMIC_EXECUTION]: The skill facilitates the execution of arbitrary PHP code and shell commands on remote Laravel Cloud environments, which is a powerful management capability.
Evidence: Features like cloud tinker {environment} --code='...' and cloud command:run {environment} --cmd='...' enable direct remote execution on the user's infrastructure (SKILL.md).
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources and remote command outputs, creating a surface for potential indirect prompt injection attacks.
Ingestion points: Fetches official documentation from https://cloud.laravel.com/docs/llms.txt and processes output from cloud deploy:monitor, cloud deployment:logs, and cloud command:get (SKILL.md).
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat external content as untrusted data.
Capability inventory: The skill possesses significant capabilities, including file writing, network fetching, and remote command execution, which could be exploited if malicious instructions are processed.
Sanitization: The skill does not describe specific sanitization or filtering of external data before interpolation into the agent's context.
[EXTERNAL_DOWNLOADS]: The skill references and installs external resources from trusted sources as part of its standard workflow.
Evidence: Fetches documentation from Laravel's official domain (cloud.laravel.com) and installs packages from the Composer registry (SKILL.md).

deploying-to-cloud