The Agent Skills Directory

[SAFE]: The skill is a collection of static markdown references providing SQL best practices. No executable code or suspicious automation was found.
[PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety filters, or extract system prompts were detected. The instructions are strictly limited to database optimization guidance.
[DATA_EXFILTRATION]: No network operations or sensitive file access patterns were found. The skill only references well-known and trusted official documentation from postgresql.org and supabase.com.
[CREDENTIALS_UNSAFE]: No hardcoded credentials were found. Placeholder text like 'xxx' is used in SQL examples for passwords, which is standard practice.
[OBFUSCATION]: No base64 encoding, zero-width characters, or homoglyph attacks were detected in URLs or scripts.
[REMOTE_CODE_EXECUTION]: The skill does not perform remote script downloads or execute unverified packages. It only provides static SQL examples for educational and optimization purposes.
[DYNAMIC_EXECUTION]: While the skill contains PL/pgSQL 'do' blocks and function definitions, these are provided as pedagogical examples for database administrators and do not represent a runtime risk to the agent host.
[SECURITY_BEST_PRACTICES]: The skill actively promotes security best practices, including the Principle of Least Privilege and the secure configuration of Row-Level Security (RLS) policies with empty search_path settings to prevent hijacking.

supabase-postgres-best-practices