My skill

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed OpenClaw utility plugin whose main risk is user-directed MCP HTTP calls to URLs the user provides.

Install only if you want these utility tools exposed to the agent. Use the MCP tools only with servers you trust, avoid localhost/internal admin or metadata endpoints unless intentional, and do not pass sensitive prompts, secrets, or workspace data to remote MCP tools unless you mean to share them.

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: 89% confidence
Finding: The MCP HTTP client sends JSON-RPC requests, including tool arguments, to arbitrary user-supplied URLs with no allowlist, trust boundary warning, or destination validation. This can expose sensitive prompts, secrets, or workspace-derived data to attacker-controlled endpoints and also enables SSRF-style access to internal services if the host environment has network reachability.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal