ClawHub
Back to skill

Security audit

Super Lap Agentos Api V3 Diary Call Group

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed agentOS diary API reference skill; it can guide live appointment changes, but I found no hidden execution, persistence, or exfiltration behavior.

Install only if you intend to let an agent work with live agentOS diary data. Use a least-privilege API key, verify tenant/branch/appointment IDs before writes, and require explicit approval before creating, updating, deleting, or canceling appointments.

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
Findings (1)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill documents destructive operations such as DELETE and cancel/update appointment actions without any warning, confirmation guidance, or guardrails about their effect on live diary data. In an agent skill context, this increases the chance that an agent or user will invoke state-changing endpoints on production systems unintentionally, causing appointment loss, cancellation, or workflow disruption.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.