The Agent Skills Directory

[SAFE]: No security issues were detected. The skill consists of static documentation files and SQL examples designed to provide guidance on database performance and security.
[INDIRECT_PROMPT_INJECTION]: The skill serves as a reference for agents. It includes robust boundary markers and warnings (particularly in AGENTS.md and SKILL.md) that explicitly instruct the agent to avoid autonomous execution of DDL or destructive commands without environment-specific validation and user approval.
[COMMAND_EXECUTION]: While the README.md and rules/ files mention shell commands (npm, cp) and SQL execution, these are presented as developer documentation and educational snippets rather than instructions for the agent to execute silently at runtime.
[EXTERNAL_DOWNLOADS]: The skill references official documentation and repositories from Supabase and the PostgreSQL Global Development Group. These are well-known, trusted sources and do not involve the download of untrusted or unverified code into the agent's runtime environment.

postgres-best-practices