The Agent Skills Directory

[SAFE]: The skill provides documentation and architectural patterns for multi-tenant systems. It does not contain any malicious code or instructions.
[PROMPT_INJECTION]: No prompt injection attempts, role-play instructions, or safety bypasses were detected in the instructions.
[CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The code snippets use placeholders like DATABASE_URL and REDIS_URL which follow standard security practices.
[COMMAND_EXECUTION]: There are no shell commands or suspicious subprocess calls. The code examples demonstrate legitimate database and cache operations through established libraries.
[DATA_EXFILTRATION]: No exfiltration patterns were identified. On the contrary, the skill explicitly provides patterns to prevent cross-tenant data leaks.
[OBFUSCATION]: The content is clear and uses standard Markdown and programming syntax. No hidden text, Base64-encoded instructions, or homoglyph attacks were detected.

multi-tenancy-patterns