ClawHub
Back to skill

Security audit

Crypto Self-Learning

Security checks across malware telemetry and agentic risk

Overview

This skill appears to keep crypto trade analysis local and update learning notes, with no evidence of theft, remote control, or destructive behavior.

Install only if you want an agent to keep local crypto trade records and persist learned rules. Review what it writes to MEMORY.md, keep trade logs free of secrets or exchange credentials, and prefer running memory updates only after explicit user approval.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly instructs the agent to read and write local files, including trade logs and MEMORY.md, but it declares no permissions in the metadata. That mismatch is a real security issue because operators and policy systems may assume the skill is less capable than it actually is, reducing transparency and weakening enforcement around file access.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill advertises automatic updates to MEMORY.md and says it appends learned rules, but it does not present a prominent warning that executing this action will modify an existing file. In an agent setting, silent or under-signaled file modification is risky because it can overwrite, poison, or persist misleading instructions into memory that later workflows may trust.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.