The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive reference for Drizzle ORM patterns and does not contain any executable code or malicious instructions that deviate from its stated purpose.\n- [CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or credentials were detected. Database connection strings are handled via standard environment variables (e.g., process.env.DATABASE_URL), which is a recommended security practice for managing secrets.\n- [REMOTE_CODE_EXECUTION]: All referenced commands (such as npx drizzle-kit) and code snippets represent standard developer workflows for the Drizzle ecosystem. No suspicious remote scripts, unverified package downloads, or piped bash executions are present.\n- [PROMPT_INJECTION]: The instructions focus purely on technical documentation and do not include patterns typical of prompt injection, such as attempts to bypass AI safety constraints, role-play as an unrestricted agent, or override core instructions.\n- [DATA_EXFILTRATION]: No network operations to untrusted domains or logic for extracting sensitive local files were found. All database operations described are standard CRUD patterns within a type-safe ORM context.

drizzle-orm-patterns