postgresql-performance
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious instructions, hidden code, or dangerous command execution patterns were identified in the skill files.
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for analyzing query execution plans and database performance statistics, which are common diagnostic data sources.
- Ingestion points: SQL queries and performance metrics fetched from the database or provided by the user (SKILL.md).
- Boundary markers: None identified, as the skill serves as a static reference for optimization patterns.
- Capability inventory: The included validation script (scripts/validate.py) is restricted to local file path checks and contains no network or broad file system capabilities.
- Sanitization: The skill employs yaml.safe_load() for configuration processing, mitigating risks associated with untrusted data deserialization.
Audit Metadata