supabase-postgres-best-practices

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Trusted Source (SAFE): The skill is attributed to Supabase and references official documentation. Per [TRUST-SCOPE-RULE], these sources are considered reliable.
  • Prompt Injection (SAFE): No patterns of instruction overriding, role-play jailbreaks, or system prompt extraction were found across the 38 files.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials or secrets were detected; SQL examples use generic placeholders like 'xxx'. No network exfiltration commands (curl, wget) or sensitive file path access (e.g., ~/.ssh) are present.
  • Obfuscation (SAFE): All content is provided in clear Markdown and SQL. No Base64 encoding, zero-width characters, or homoglyph-based evasion techniques were identified.
  • Code Execution (SAFE): The skill contains documentation and static SQL examples. While it mentions an npm build process for validation, no suspicious remote code execution or piped bash scripts are present.
  • Indirect Prompt Injection (LOW): (1) Ingestion point: The agent reads various Markdown files in the references directory. (2) Boundary markers: Content is clearly delimited by Markdown headers and code blocks. (3) Capability inventory: The skill is intended to guide SQL generation and optimization. (4) Sanitization: Content is static and provided by a trusted source, presenting a low surface area for adversarial input injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:42 PM