The Agent Skills Directory

[SAFE]: The skill serves as a best-practice manual for verifying Row Level Security implementation. It focuses on identifying known security pitfalls such as superuser bypass, table owner bypass, and insecure views. All external links point to authoritative documentation or well-known technical blogs.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external, potentially untrusted code files such as SQL migrations and application source code. This creates a surface for indirect prompt injection if those files contain malicious instructions. * Ingestion points: Analyzes files matching patterns like **/migrations/**/*, **/*policy*.sql, and **/auth/**. * Boundary markers: The skill encourages the use of a structured RLS implementation artifact for reporting results, which provides a level of output scoping. * Capability inventory: Performs code analysis and generates implementation summary artifacts. * Sanitization: The skill does not explicitly define sanitization steps for the data it ingests, relying on the agent's logic to process the SQL and code files.

postgres-rls