ClawHub
Back to skill

Security audit

personal-context

Security checks across malware telemetry and agentic risk

Overview

This instruction-only skill stores a local personal profile for personalization, with privacy-relevant persistence that is disclosed and aligned with its purpose.

Install this only if you want OpenClaw to remember basic personal preferences across sessions. Review or edit ~/.openclaw/workspace/me.json after onboarding, and do not provide passwords, tokens, financial details, health information, or other sensitive data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill's activation rules are broad enough to trigger on ambiguous phrases and inferred conditions like a 'first session' or when the agent thinks its tone feels off. That can cause unsolicited collection of personal data and profile creation without a clearly intentional user request, creating privacy and consent risks even if no overtly sensitive data is requested.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The instruction to 'silently load' a personal profile every session means the agent reuses stored personal data without a contemporaneous user-facing notice or clear reminder. In a personalization skill, this increases the chance users are unaware of ongoing access to identity and preference data, which weakens transparency and informed consent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.