ClawHub
Back to skill

Security audit

RugCheck

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a read-only Solana token risk checker that calls the RugCheck API, with no evidence of credential access, persistence, exfiltration, or destructive behavior.

Install only if you are comfortable with the agent running a local shell helper that sends Solana token mint addresses to RugCheck. Treat its results as third-party risk signals, not financial advice, and prefer explicit token-checking requests when invoking it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill instructs use of a shell script (`bash scripts/rugcheck.sh ...`) but does not declare any permissions for shell execution. Undeclared execution capability can bypass user/operator expectations and, if the script later includes unsafe argument handling or network/file operations, the agent may execute commands with more power than its manifest suggests. In this context the shell is used for read-oriented token lookups, which lowers severity somewhat, but the hidden capability still meaningfully increases risk.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger language is broad enough to activate on general token discovery, trending, verification, or vague 'is this safe' requests rather than only explicit RugCheck analysis requests. Over-broad routing can cause the agent to invoke this skill unexpectedly, making external API calls or shell commands in situations the user did not clearly request and potentially crowding out safer or more appropriate skills. The context is somewhat less dangerous because the skill is informational and focused on Solana tokens, but ambiguous activation still creates unnecessary attack surface and consent/routing risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.