ClawHub
Back to skill

Security audit

Atlas Framework

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only development framework skill with disclosed local project memory setup, not hidden or malicious behavior.

Install if you want a structured AI-assisted development workflow. Before first use, confirm where the memory and SQLite files will be created, keep credentials and sensitive client or personal data out of memory/logs, and periodically inspect or delete those files if you no longer need them.

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
95% confidence
Finding
The skill explicitly directs the agent to read and create persistent memory files and initialize activity-tracking databases, but it does not require user consent, disclosure, retention limits, or guidance on handling sensitive data. In an agent framework context, this can lead to unexpected long-term storage of user prompts, secrets, behavioral data, or workspace contents, creating privacy, compliance, and data minimization risks.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.