ClawHub

Grill Gate

Security checks across malware telemetry and agentic risk

Overview

This skill does what it says: it installs an OpenClaw runtime gate that can block selected agent tool calls until a local review token is present.

Install only if you want a plugin that can block agent exec or subagent-spawn actions. Review and narrow the trigger words, blocked commands, token directory, and token TTL before relying on it, and do not assume tokens are truly one-use unless the implementation is fixed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger list includes very generic terms like 'design', 'plan', 'develop', and 'research', which are likely to appear in many ordinary prompts. That can cause overblocking or policy manipulation, where benign tasks are unnecessarily forced through the grill workflow, creating denial-of-service against normal agent operation and incentivizing users to adopt unsafe exemptions.

Session Persistence

Medium
Category
Rogue Agent
Content
## Configuration

Create `~/.openclaw/grill-gate.json` (all fields optional):

```json
{
Confidence
77% confidence
Finding
Create `~/.openclaw

Unpinned Dependencies

Low
Category
Supply Chain
Content
"typescript": "^6.0.3"
  },
  "dependencies": {
    "openclaw": "^2026.5.22"
  }
}
Confidence
84% confidence
Finding
"openclaw": "^2026.5.22"

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal