zoom-shell-integration
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions explicitly promote secure development practices for embedded systems, such as configuring bounded buffers to prevent overflows and implementing guards for dangerous commands like firmware updates or data erasure.
- [SAFE]: No evidence of prompt injection or attempts to bypass AI safety guidelines was found. The instructions are clearly scoped to MCU firmware development.
- [SAFE]: No suspicious data access, exfiltration patterns, or hardcoded credentials were detected. The skill operates within the context of user-provided technical information.
- [SAFE]: The skill does not perform any external network operations, remote code downloads, or unauthorized package installations.
- [SAFE]: No obfuscation techniques, hidden payloads, or persistence mechanisms are present in the provided content.
Audit Metadata