Stuck Trace

Security checks across malware telemetry and agentic risk

Overview

This prompt-only skill is privacy-sensitive because it reads agent memory files, but that behavior is disclosed, purpose-aligned, and bounded by output privacy rules.

Before installing, understand that this skill is designed to let the agent read profile, memory, and recent conversation files when you ask it to analyze why a project or collaboration is stuck. Use it only in an environment where you are comfortable with that memory access, and avoid using it around highly sensitive personal, medical, financial, or workplace-confidential records unless the host agent gives you adequate control over what files are available.

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: 98% confidence
Finding: The skill is explicitly designed to automatically read SOUL.md, USER.md, memory files, and recent conversations to generate output, yet it does not require a clear user-facing notice or consent step before accessing potentially sensitive personal archives. In a memory-enabled agent, this can expose private contextual data, infer sensitive attributes, or normalize silent surveillance-style data access beyond the user’s immediate prompt.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal