ClawHub

AgentOps Guardrails

Security checks across malware telemetry and agentic risk

Overview

This skill is a local incident-detection and routing toolkit with disclosed file reads and local state writes, and no evidence of hidden network access or credential handling.

Install only if you want a local agent-ops detector/router that can read configured runtime files and write incident outputs. Start with dry-run mode, choose workspace and state-file paths deliberately, review implicit invocation before enabling broad use, and run clean-generated.sh only when you are comfortable deleting generated .jsonl and .lock files under the skill folder.

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
93% confidence
Finding
The skill invokes shell scripts that read from and write to local files, but the manifest does not declare any permissions for those capabilities. This creates a transparency and policy-enforcement gap: a caller may treat the skill as low-privilege while it can actually modify workspace state, incident files, and lock artifacts through bash tooling.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal