The Agent Skills Directory

[SAFE]: The skill provides instructional content and code snippets for configuring and deploying Craft CMS applications on the Craft Cloud platform. All documented behaviors, including the Build → Migrate → Release pipeline and the use of the craftcms/cloud extension, are standard operational procedures for this serverless environment.
[SAFE]: Security best practices are emphasized throughout the documentation. For example, the skill provides specific warnings against hardcoding credentials in db.php, correctly identifies the risks of CSRF token leakage in cached templates, and provides remediation patterns using the csrfInput() function and async CSRF handling.
[SAFE]: External dependencies and references are limited to official documentation (craftcms.com), trusted source repositories (github.com/craftcms), and well-known open-source libraries (e.g., ky, http-message-sig). These are relevant to the skill's primary purpose and originate from reputable sources.
[SAFE]: Guidance on secret management aligns with industry standards, instructing users to manage sensitive environment variables through the platform's Console UI rather than committing them to version control or local .env files.
[SAFE]: No evidence of prompt injection, obfuscation, privilege escalation, or persistence mechanisms was found. The instructions are clearly defined and purpose-built for AI agents helping with Craft Cloud infrastructure.

craft-cloud