database-optimization-guru

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection.
  • Ingestion points: Ingests user-provided SQL schemas, queries, and performance logs as described in modules/core-guidance.md and modules/workload-profiling.md.
  • Boundary markers: The instructions do not specify the use of delimiters or explicit "ignore instructions" for the user-provided content.
  • Capability inventory: The skill recommends utilizing performance tools like pgBench and sysbench in modules/workload-profiling.md, and includes functionality to import documentation from external repositories.
  • Sanitization: Documentation in modules/workload-profiling.md correctly advises scrubbing sensitive data when preparing benchmark environments.
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches content from external repositories and services.
  • Trusted Sources: Recommends importing documentation from the official postgres/postgres GitHub repository using the adn_skills tool in modules/research-checklist.md.
  • Official Documentation: Provides links to official performance documentation and manuals for PostgreSQL, MySQL, AWS, Google Cloud, and Azure SQL.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 05:38 PM
Security Audit — agent-trust-hub — database-optimization-guru