specialist-engineering-database-optimizer

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill interpolates user-provided content directly into the instructions for a sub-agent. This represents an indirect prompt injection surface where a malicious user could attempt to influence the sub-agent's behavior through their query.
  • Ingestion points: {the user's question} variable in SKILL.md.
  • Boundary markers: None present to isolate user input from the specialist's system instructions.
  • Capability inventory: Spawns ephemeral specialist agents via the Agent tool.
  • Sanitization: No escaping or validation of the input string is performed before interpolation.
  • [SAFE]: The skill includes code snippets referencing the official @supabase/supabase-js library, which is a well-known service.
  • [SAFE]: The skill demonstrates safe credential management by using process.env for sensitive tokens and connection strings rather than hardcoding values.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 08:49 PM
Security Audit — agent-trust-hub — specialist-engineering-database-optimizer