sql-optimizer

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes external SQL queries and EXPLAIN plans as untrusted input (SKILL.md). It does not incorporate boundary markers or clear instructions to treat the provided query text as non-executable data, creating a potential surface for indirect prompt injection if an attacker embeds agent instructions within SQL comments or strings.
  • Ingestion points: User-supplied SQL queries and EXPLAIN plan outputs.
  • Boundary markers: None identified in the workflow or output templates.
  • Capability inventory: No tools are explicitly requested in the skill manifest, which significantly limits the potential impact of an injection.
  • Sanitization: There are no instructions for sanitizing or escaping the input queries before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 06:51 PM