convex-backend

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions found that attempt to override agent safety or system instructions.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded secrets, API keys, or access to sensitive local file paths detected.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): No automated package installation (npm/pip) or remote script execution (curl/wget) commands are present. The link to docs.convex.dev is for reference only.
  • [Persistence & Privilege Escalation] (SAFE): No attempts to modify system configurations, shell profiles, or acquire elevated privileges.
  • [Obfuscation] (SAFE): The content is clear and uses standard Markdown/TypeScript syntax with no hidden characters or encoding.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:24 PM