The Agent Skills Directory

[SAFE]: No malicious patterns, command injections, or unauthorized network operations were detected. The instructions are focused on providing expert guidance for database performance tuning.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface as it is designed to analyze database query logs and statistics which could contain untrusted data. 1. Ingestion points: Shared statistics and query logs as mentioned in the Measurement and change scope section. 2. Boundary markers: The instructions recommend using approved sandboxes and maintenance windows with timeouts and requiring authorization for consequential changes. 3. Capability inventory: The agent is guided to use commands like EXPLAIN ANALYZE and perform schema modifications such as index rebuilding or migrations. 4. Sanitization: The skill lacks explicit instructions for sanitizing instructions that might be embedded within the logs or query strings, though it warns about the exposure of sensitive values in logs.
[NO_CODE]: This skill consists entirely of documentation and instructions in markdown format with no associated executable scripts or binaries.

database-optimizer