The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute the psql command-line tool. This includes running arbitrary SQL queries via the -c flag and executing SQL scripts from files using the -f flag. This is a primary and intended function of the skill.
[EXTERNAL_DOWNLOADS]: The skill metadata contains installation instructions for the postgresql-client using system package managers including Homebrew (macOS), APT (Linux), and Scoop (Windows). These are well-known, trusted package management systems and are used to fulfill the skill's requirements.
[SAFE]: The skill uses industry-standard practices for database connection management, specifically recommending the use of environment variables (e.g., DATABASE_URL) to handle sensitive connection strings rather than hardcoding credentials.
[INDIRECT_PROMPT_INJECTION_SURFACE]:
Ingestion points: The skill reads data from external PostgreSQL databases through query results returned to the psql CLI (SKILL.md).
Boundary markers: No explicit delimiters are used in the provided examples to wrap database output.
Capability inventory: The skill can execute shell commands (psql) and interact with file systems (psql -f).
Sanitization: No explicit sanitization of database content is performed within the instructions; the agent relies on its own processing logic for the returned data.

postgres