OpenClaw Performance

Security checks across malware telemetry and agentic risk

Overview

This is a local performance-analysis skill that reads user-selected source files and does not show evidence of network transfer, credential use, persistence, or destructive behavior.

Install only if you are comfortable letting the tool read the source files or directories you pass to it. Prefer running it on explicit project paths, and be careful with any profiling or benchmarking commands that would execute untrusted code.

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 (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill clearly instructs users to analyze files and directories (for example `--file` and `--dir` inputs), which implies file-reading capability, but the documentation does not declare permissions. Undeclared file access weakens reviewability and consent boundaries, making it easier for a skill to read unintended local source files during analysis.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal