The Agent Skills Directory

[COMMAND_EXECUTION]: The skill references standard command-line tools for database management, such as npx prisma and psql. These are used appropriately within the context of database migrations and schema validation in development and CI/CD environments.
[CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The skill correctly recommends using environment variables (e.g., env("DATABASE_URL")) for sensitive configuration, which is an industry-standard security practice.
[OBFUSCATION]: A thorough review of all instructions and code snippets revealed no signs of obfuscation, including Base64 encoding, zero-width characters, or homoglyph attacks. The content is presented in clear, plain text.
[REMOTE_CODE_EXECUTION]: The skill does not contain any patterns for downloading or executing scripts from untrusted remote sources. It relies on well-known, established tools like Prisma.
[DATA_EXFILTRATION]: No network operations to untrusted domains or attempts to access sensitive system files (e.g., SSH keys, AWS credentials) were detected.
[PROMPT_INJECTION]: The instructions follow standard technical guidance formats and do not include attempts to override agent behavior, bypass safety filters, or extract system prompts.
[INDIRECT_PROMPT_INJECTION]: The skill focuses on providing static guidelines for database architecture. It does not possess an attack surface for ingesting untrusted external data that could influence agent logic.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not utilize dynamic context patterns (e.g., shell commands executed at load time) to perform operations or exfiltrate data.

db-enforcer