deploying-to-cloud
Warn
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: MEDIUMREMOTE_CODE_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill facilitates arbitrary shell command execution on remote Laravel Cloud environments through the
cloud command:runcommand. - [DYNAMIC_EXECUTION]: The skill enables the execution of dynamic PHP code on remote infrastructure via the
cloud tinkerutility. - [INDIRECT_PROMPT_INJECTION]: The skill ingests potentially untrusted data from external sources, which could contain malicious instructions designed to influence agent behavior.
- Ingestion points: Documentation fetched from
https://cloud.laravel.com/docs/llms.txtand deployment logs retrieved viacloud deployment:logs. - Boundary markers: There are no explicit delimiters or instructions provided to isolate ingested data from the agent's core instructions.
- Capability inventory: The skill has high-privilege remote execution capabilities including shell access and PHP evaluation.
- Sanitization: The instructions do not specify any sanitization or filtering of the external content before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill uses the
cloudCLI to perform various management tasks, including the creation and deletion of infrastructure and secrets. - [EXTERNAL_DOWNLOADS]: The skill instructs the user to install several packages via Composer, including
laravel/cloud-cli,resend/resend-php, andleague/flysystem-aws-s3-v3. These are identified as vendor-owned or well-known resources relevant to the skill's stated purpose.
Audit Metadata