nutcrackertestgpt

Security checks across malware telemetry and agentic risk

Overview

This local UX research skill is not malicious, but it needs review because it can read and retain OpenClaw conversation histories more broadly than its consent prompt clearly states.

Install only if you are comfortable with local analysis of your OpenClaw conversations. Prefer this_session_only and minimal or snippets capture, avoid all-agent collection unless you explicitly want it, review generated reports before sharing them, and use purge or a short retention window for sensitive work.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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: 83% confidence
Finding: The template explicitly includes filesystem paths to raw event logs and session indexes in the generated report, but provides no warning that these artifacts may contain sensitive conversation content, local usage traces, or personal data. In a privacy-first ethnography skill, surfacing these locations in a report increases the chance of accidental disclosure, wider internal sharing, or retention beyond intended scope.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal