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