The Agent Skills Directory

[SAFE]: The skill serves as a reference guide and advisor for PostgreSQL database tuning, utilizing standard diagnostic queries and optimization techniques such as EXPLAIN ANALYZE and autovacuum tuning.
[DATA_EXPOSURE]: No hardcoded credentials or sensitive environment data were found; example connection strings and pgbouncer configurations use generic placeholders like 'user:pass' and 'localhost'.
[INDIRECT_PROMPT_INJECTION]: The skill contains logic specifically designed to validate user-provided SQL against security anti-patterns like string concatenation, effectively acting as a security linter for database queries. Ingestion points: User-provided SQL and application code snippets evaluated against the reference system. Boundary markers: Explicit instructions in SKILL.md to ground responses in provided reference files. Capability inventory: The skill provides text-based advice and diagnostic query generation without direct shell execution or network request capabilities. Sanitization: Implements detailed regex-based validation rules in references/validations.md to identify and warn against unsafe query patterns.

postgres-wizard