Aport Agent Guardrail
Security checks across malware telemetry and agentic risk
Overview
This appears to be a legitimate guardrail skill, but it persistently changes agent tool execution and has cloud-mode/data-handling ambiguity that users should review before installing.
Install only if you want a persistent OpenClaw policy-enforcement hook. During setup, explicitly choose local mode if you do not want tool-call metadata sent to APort, review the generated ~/.openclaw config and wrapper scripts, and confirm how to disable or reset the plugin before relying on it in a sensitive environment.
SkillSpector
By NVIDIA
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
VirusTotal
64/64 vendors flagged this skill as clean.